molecular studies of the pathogenic free-living amoeba, acanthamoeba

Molecular Studies of the Pathogenic Free-living

Amoeba, Acanthamoeba

Thesis submitted for the degree of

Doctor of Philosophy

at the University of Leicester

by

Kimberley Amy Durham B.Sc. M.Sc.

Department of Infection, Immunity & Inflammation

University of Leicester

April 2012

ii

Abstract

Molecular Studies of the Pathogenic Free-living Amoeba, Acanthamoeba

By Kimberley A. Durham

Ubiquitous amoebae from the genus Acanthamoeba are associated with two

main serious infections: The more common eye disease acanthamoeba keratitis

(AK), which can result in blindness, and the rare and often fatal disease affecting

the central nervous system, granulomatous amoebic encephalitis (GAE).

The traditional morphological taxonomic system for Acanthamoeba is based

on cyst size and shape, and divides the amoebae into three groups (I, II and III).

Since the discovery that cyst shape can be modified by culture conditions, the

classic system has become largely redundant. A more robust system has been

developed, based on the nucleotide sequence of the 18S rRNA (Rns) gene. It types

Acanthamoeba into 15 T-groups, with most species including environmental and

clinical clumped into three groups T3, T4 and T11, with little resolution between

them. Although speciation does not help cure patients directly, it can provide

valuable information regarding disease epidemiology and ultimately benefit patient

prognosis.

Here a system to better resolve strains has been developed, using the

mitochondrial cytochrome oxidase subunit 1 and 2 (cox1/2) gene sequence. When

used in conjunction with the T-group system, resolution between strains including

those with a T3, T4 or T11 genotype is obtained. Additionally the combined

approach identified a mixed infection in a patient suffering with AK, and the

occurrence of Acanthamoeba strains with multiple alleles of 18S and cox1/2 genes.

The combined use of both genotyping systems was used to investigate an

unprecedented outbreak of GAE within a Swedish hospital. Results confirmed

Acanthamoeba had infected several immunocompromised paediatrics from a single

ICU, and the source was from within the unit’s water system. In vitro assays were

used to test the strains pathogenic abilities and sensitivities to antimicrobial

compounds, identifying if they are more virulent than typical strains of

Acanthamoeba.

iii

Acknowledgements

This work was supported by sponsorship funding from Bausch and Lomb. I

also owe thanks to most of the members of the Department of Infection, Immunity

and Inflammation for their assistance throughout my research, in particular Dr

Wayne Heaselgrave, Dr James Lonnen, Niran Patel and Professor Peter Andrew.

I would like to give a special thanks to my family for their continued support

without which this write up would not have possible.

iv

Table of contents

1 INTRODUCTION ............................................................................................ 1

1.1 Free-living amoebae ................................................................................... 1

1.2 Acanthamoeba ............................................................................................ 5

1.3 Acanthamoeba ecology ............................................................................... 7

1.4 Acanthamoeba cultivation ........................................................................ 10

1.5 Acanthamoeba epidemiology .................................................................... 12

1.6 Granulomatous amoebic encephalitis (GAE) ........................................... 15

1.7 Acanthamoeba keratitis (AK) .................................................................... 19

1.8 Taxonomy and Classification.................................................................... 24

1.9 Molecular biology of Acanthamoeba ........................................................ 28

1.10 Aims .......................................................................................................... 33

2 DNA TYPING OF ACANTHAMOEBA SP. ................................................. 34

2.1 Introduction ..................................................................................................... 34

2.1.1 Aims .................................................................................................. 37

2.2 Materials and Methods ................................................................................... 38

2.2.1 Chemicals .................................................................................................. 38

2.2.2 Organisms ................................................................................................. 38

2.2.3 Monoxenic culture of Acanthamoeba ....................................................... 40

2.2.3.1 Preparation of E. coli food source stock ........................................... 41

v

2.2.4 Axenic culture of Acanthamoeba .............................................................. 42

2.2.5 Cryopreservation of Acanthamoeba ......................................................... 43

2.2.5.1 Cryopreservation of bacteria ............................................................. 45

2.2.6 Acanthamoeba DNA isolation .................................................................. 45

2.2.6.1 DNA extraction from Acanthamoeba in monoxenic culture ............ 46

2.2.7 Agarose gel electrophoresis ...................................................................... 47

2.2.8 Primer design ............................................................................................ 48

2.2.9 PCR amplification of DNA ....................................................................... 48

2.2.10 Purification of PCR products .............................................................. 49

2.2.10.1 PEG purification of single bands ...................................................... 50

2.2.10.1.i Micron®–PCR purification of single bands 50

2.2.10.2 Target DNA purification from a multiple band PCR ....................... 51

2.2.11 Ligation of DNA into cloning vectors .................................................. 52

2.2.12 Production of ultra competent E. coli cells ......................................... 53

2.2.13 Heat shock transformation of ultra competent E. coli ......................... 54

2.2.14 Restriction enzyme digestion ............................................................... 55

2.2.15 Plasmid purification ........................................................................... 58

2.2.15.1 Silica method of plasmid purification for screening ......................... 58

2.2.15.2 Comprehensive plasmid purification for sequencing ....................... 60

2.2.15.3 QIAGEN® plasmid maxi kit ............................................................ 61

2.2.16 Sequencing ........................................................................................... 61

2.2.17 Sequence analysis ................................................................................ 62

vi

2.3 Results .............................................................................................................. 63

2.3.1 18S genotyping of strains .......................................................................... 63

2.3.2 Typing strains with cox1/2 gene ............................................................... 69

2.3.3 Introns, multiple alleles and mixed infections .......................................... 77

2.3.4 Concatenated sequence data .................................................................... 78

2.4 Discussion ........................................................................................................ 84

2.4.1 Genotyping Acanthamoeba sp. with 18S sequences ................................. 84

2.4.2 Genotyping based on cox1/2 sequences .................................................... 87

2.4.3 Typing with concatenated gene sequences ............................................... 88

2.4.4 Multiple alleles and mixed infections using 18S and cox1/2 .................... 89

2.4.5 Conclusion ................................................................................................ 91

3 SWEDISH GAE STUDY ................................................................................ 93

3.1 Introduction ..................................................................................................... 93

3.1.1 Aims .................................................................................................. 95

3.2 Materials and Methods ................................................................................... 96

3.2.1 Establishing amoebae cultures ................................................................. 96

3.2.2 Amoebae DNA extraction ......................................................................... 97

3.2.2.i Direct DNA extraction from CSF 97

3.2.3 DNA manipulation .................................................................................... 98

3.2.4 Temperature tolerance assay .................................................................... 98

vii

3.2.5 Osmotolerance assay ................................................................................ 99

3.2.6 Protease secretion ..................................................................................... 99

3.2.7 Complement fixing potential ................................................................... 100

3.2.8 Cytopathogenic potential ........................................................................ 101

3.2.9 Antimicrobial sensitivity assays .............................................................. 102

3.2.9.1 Trophozoites ................................................................................... 102

3.2.9.2 Cysts ................................................................................................ 103

3.3 Results ............................................................................................................ 106

3.3.1 Culture and morphological analysis....................................................... 106

3.3.1.1 Patient one ....................................................................................... 106

3.3.1.2 Patient two ...................................................................................... 106

3.3.1.3 Patient three .................................................................................... 107

3.3.1.4 Hospital water samples ................................................................... 107

3.3.2 Molecular analysis .................................................................................. 109

3.3.2.1 18S phylogenetic analysis ............................................................... 109

3.3.2.2 Cox1/2 phylogenetic analysis ......................................................... 113

3.3.3 Pathogenicity assays ............................................................................... 115

3.3.3.1 Amoebicidal activity of human serum ............................................ 115

3.3.3.2 Tolerance of increased temperature and osmolarity ....................... 116

3.3.3.3 Protease secretion assays ................................................................ 116

3.3.3.4 Cytopathogenicity of Acanthamoeba .............................................. 117

3.3.4 Drug sensitivity assays ............................................................................ 120

3.4 Discussion ...................................................................................................... 124

viii

3.4.1 Molecular ................................................................................................ 124

3.4.2 Pathogenicity .......................................................................................... 127

3.4.3 Antimicrobial sensitivity ......................................................................... 131

3.4.1 Conclusion .............................................................................................. 136

4 REFERENCES .............................................................................................. 138

ix

List of Figures

Figure 1. Neighbour joining distance tree based on partial 18S rDNA sequences of

Acanthamoeba spp. The tree is unrooted and obtained by Kimura two-parameters

correction for multiple substitutions using MEGA (5.05). The tree is based on

reference bp from 1,175 to 1,379. The scale bar represents the corrected number of

nucleotide substitutions per base using Kimura method. Designated T-groups are

shown. ....................................................................................................................... 67

Figure 2. Maximum parsimony distance tree based on partial 18S rDNA sequences

of Acanthamoeba spp. The tree is unrooted and obtained by Kimura two-parameters


reference bp from 1,175 to 1,379. Designated T-groups are shown. ........................ 69

Figure 3. Maximum parsimony tree based on partial mt cox1/2 sequences of

Acanthamoeba spp, with cox1/2 groups (A-H) and T-groups (T2-T11) included. The

tree is unrooted and obtained by Kimura two-parameters correction for multiple

substitutions using MEGA (5.05). The tree is based on reference bp 8,002 to 8,566.

Bootstrap values have been included, based on 1,000 bootstrap values, and are

placed at the nodes they apply to. ............................................................................. 74

Figure 4. Neighbour-joining distance tree based on partial mt cox1/2 sequences of

Acanthamoeba spp with cox1/2 groups (A-H) and T-groups (T2-T11) included. The



The scale bar represents the corrected number of nucleotide substitutions per base

using Kimura method. ............................................................................................... 76

x

Figure 5. Maximum parsimony distance tree based on concatenated partial mt

cox1/2 with 18S sequences of Acanthamoeba spp with cox1/2 groups (A-H) and T-

groups (T2-T11) included. The tree is unrooted and obtained by Kimura two-

parameters correction for multiple substitutions using MEGA (5.05). The tree is

based on reference bp 8,002 to 8,566. ...................................................................... 81

Figure 6. Neighbour-joining distance tree based on concatenated partial mt cox1/2

with 18S sequences of Acanthamoeba spp with cox1/2 groups (A-H) and T-groups

(T2-T11) included. The tree is unrooted and obtained by Kimura two-parameters


reference bp 8,002 to 8,566. The scale bar represents the corrected number of

nucleotide substitutions per base using Kimura method. ......................................... 83


of Swedish hospital FLA with comparison species. The tree is unrooted and

obtained by Kimura two-parameters correction for multiple substitutions using

MEGA (5.05). Analysis is based on reference bp from 1,175 to 1,379. Designated

T-groups and strain origins are shown. Bootstrap values have been included, based

on 1,000 bootstrap values, and placed at the corresponding nodes. ....................... 112

Figure 8. Maximum parsimony tree based on partial mitochondrial cox1/2

sequences of Acanthamoeba spp. The tree is unrooted and obtained by Kimura two-


based on reference bp 8,002 to 8,566. Designated T-groups and origin of strains are

shown. Bootstrap values have been included, based on 1,000 bootstrap values, and

placed at the corresponding nodes. ......................................................................... 114

xi






placed at the corresponding nodes. The scale bar represents the corrected number of

nucleotide substitutions per base using Kimura method. ....................................... 115

Figure 10. Activity of Normal Human Serum against Acanthamoeba trophozoites.

A, effect of complement from NHS on Acanthamoeba trophozoites; B, control,

Acanthamoeba incubated with heat-inactivated NHS; C, pellet formed in the round

bottomed well after lysis of Acanthamoeba trophozoites. ...................................... 115

Figure 11. Sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-

PAGE) containing gelatine, to observe the presence of extracellular proteases in

Acanthamoeba culture medium (ACM). Lane 1, ICU pool; 2, Shower room 3; 3,

Shower room 15; 4, A. polyphaga Ros; 5, Patient 3; 6, Patient 2; 7, Left blank; 8,

negative control, containing sterile culture media. ................................................. 117

xii

List of Tables

Table 1. Morphologically designated Acanthamoeba species assigned to T-groups

based on phylogenetic analysis of 18s sequence variations. .................................... 26

Table 2. Microorganisms used for these studies. ...................................................... 38

Table 3. Primers (5' to 3') used for gene sequence typing of Acanthamoeba spp.

SSU 18S rDNA and cox1/2, with corresponding PCR parameters. ......................... 56

Table 4. Estimates of evolutionary divergence over 18S sequence pairs between T-

group genotypes. The number of base substitutions per site from between sequences

are shown. Analyses were conducted using the Maximum Composite Likelihood

model (Tamura et al., 2004). The analysis involved 40 nucleotide sequences. All

positions containing gaps and missing data were eliminated. There were a total of

336 positions in the final dataset. Evolutionary analyses were conducted in MEGA5

(Tamura et al., 2011) in press. .................................................................................. 64

Table 5. Estimates of evolutionary divergence over cox1/2 sequence pairs between

groups (A-G). The number of base substitutions per site from between sequences






Table 6. Pairwise distance values between cox1/2 sequences of repeat isolates from

Patient three. The number of base substitutions per site from between sequences are

shown. Analyses were conducted using the Maximum Composite Likelihood model

(Tamura et al., 2004). The analysis involved 40 nucleotide sequences. All positions

xiii

containing gaps and missing data were eliminated. There were a total of 336

positions in the final dataset. Evolutionary analyses were conducted in MEGA5


Table 7. FLA isolated from clinical and hospital samples collected in a Swedish

hospital, with morphological identification of genus groups. ................................ 108

Table 8. Pathogenicity characteristics of Acanthamoeba sp. collected from an

outbreak of GAE with a single ward of a Swedish hospital, and compared with an

AK isolate and a non-pathogen. .............................................................................. 119

Table 9. In vitro sensitivities of five antimicrobial compounds against several strains

of Acanthamoeba trophozoites and cysts. ............................................................... 122

Table 10. Acanthamoeba in vitro Minimum Cysticidal Concentrations (MCC),

against five antimicrobial compounds. ................................................................... 122

xiv

Abbreviations

AK Acanthamoeba keratitis

APS Ammonium persulfate

ARB Amoeba-resistant bacteria

ATCC American Type Culture Collection

CCAP Culture Collection of Algae and Protozoa

CFU Colony forming units

CNS Central nervous system

CSF Cerebrospinal fluid

CT Computerised tomography

DH5α Escherichia coli (ATTC 53868)

DMEM Dulbecco’s Modified Eagle Medium

DMSO Dimethyl sulfoxide

dNTP Deoxynucleotides

dPBS Dulbecco’s Phosphate buffered saline

EDTA Ethylenediaminetetraacetic acid

EtOH Ethanol

FBS Foetal bovine serum

FLA Free-living amoeba

GMS Gomori methamine silver

HCl Hydrochloric acid

IIF Indirect immunofluorescence

IPTG Isopropyl-β-D-thiogalactoside

KH2PO4 Potassium dihydrogen orthophosphate

LB Luria-Bertani

xv

LSHTM London School of Hygiene & Tropical Medicine

LSU Large subunit

MgCl2 Magnesium chloride

MP Maximum parsimony

MRI Magnetic resonance imagery

MRSA Methicillin-resistant Staphylococcus aureus

NaCl Sodium chloride

NaOH Sodium hydroxide

NHS Normal human serum

nH2O Nanopure water

NJ Neighbour-joining

(NH4)2SO4 Ammonium sulphate

NNA Non-nutrient agar

PAGE Polyacrylamide gel

PAM Primary amoebic meningoencephalitis

PAS Periodic acid-Schiff

PCR Polymerase chain reaction

PEG Polyethylene glycol

PNACL Protein Nucleic Acid Chemistry Laboratory

RE Restriction enzymes

Rnl The large subunit rRNA gene

Rns The gene encoding for 18S

SDS Sodium dodecyl sulphate

SOB Super optimal broth

SOC SOB with catabolite repression (added glucose)

xvi

SSU Small subunit

TEM Transmission electron microscopy

TEMED Tetramethylethylenediamine

X-gal 5-bromo-4-chloro-3-indolyl-β-D-galactoside

xvii

Suppliers

ABGene

The Amoebae

Laboratory

Applied Biosystems

Arch U.K. Biocides

ATCC

Barloworld Scientific

Ltd

Bausch & Lomb

Thermo Scientific (ABGene), ABgene House, Blenheim

Road, Epsom, KT19 9AP, U.K.

Department of Infection, Immunity and Inflammation,

Maurice Shock Medical Sciences Building, University

Road, Leicester, LE1 9HN.

Applied Biosystems, Lingley House, 120 Birchwood

Boulevard, Warrington, WA3 7QH, U.K.

Arch U.K. Biocides Ltd, Wheldon Road, Castleford,

West Yorkshire, WF10 2JT, U.K.

American Type Culture Collection (ATCC), LGC

Standards, Queens Road, Teddington, Middlesex, TW11

0LY, U.K.

Barloworld Scientific Ltd, Stone, Staffordshire, ST15

0SA

Bausch & Lomb U.K. Ltd, 106 London Road, Kingston

upon Thames, Surrey, KT2 6TN, U.K.

xviii

BD Biosciences

CCAP

Eurofins MWG

Operon

Fisher Scientific UK

(NUNC & Nalgene)

Helena Biosciences

Invitrogen

Johnson & Johnson

Kemprotec Ltd

BD Biosciences, Edmund Halley Road, Oxford Science

Park, Oxford, OX4 4DQ, U.K.

CCAP, SAMS Research Services Ltd, Scottish Marine

Institute, Oban, Argyll, PA37 1QA, U.K.

Eurofins MWG Operon, Westway Estate, 28-32 Brunel

Road, Acton, London, W3 7XR, U.K.

Fisher Scientific UK, Bishop Meadow Road,

Loughborough, Leicestershire, LE11 5RJ, U.K.

Helena Biosciences Europe, Queensway South, Team

Valley Trading Estate, Gateshead, Tyne and Wear,

NE11 0SD, U.K.

Invitrogen Ltd, 3 Fountain Drive, Inchinnan Business

Park, Paisley, PA4 9RF, U.K.

Johnson & Johnson, 1 Johnson & Johnson Plaza, New

Brunswick, New Jersey, 08933, USA

Kemprotec Ltd, 11 Pennyman Green, Maltby,

Middlesbrough, TS8 0BX, U.K.

xix

Lab MTM

LSHTM

Lund University

Millipore

Molekula

Moorfields Eye

Hospital

Oxoid Limited

Pall Life Sciences

Lab M, Topley House, 52 Wash Lane, Bury, BL9 6AS,

U.K.

London School of Hygiene and Tropical Medicine,

Keppel Street, London, WC1E 7HT, U.K.

Lund University, Division of Medical Microbiology, Hs

32, Medicinsk Mikrobiologi, Paradisgaten 2, Lund,

Sweden.

Millipore, Units 3 & 5, The Courtyards, Hatters Lane,

Watford, WD18 8YH, U.K.

Molekula Ltd, 6A Arrow Trading Estate, Corporation

Road, Audenshaw, Manchester, M34 5LR, U.K.

Moorfields Eye Hospital NHS Foundation Trust, 162

City Road, London, EC1V 2PD, U.K.

Oxoid Ltd, Wade Road, Basingstoke, Hampshire, RG24

8PW, U.K.

Pall Life Sciences, 5 Harbourgate Business Park,

Southampton Road, Portsmouth, PO6 4BQ, U.K.

xx

PNACL

Promega UK

Puritan Medical

Products

Qiagen Ltd

Sanyo Gallenkamp

Sanofi-Aventis

Sigma-Aldrich

Syngene

PNACL, The University of Leicester, University Road,

Leicester, LE1 7RH, U.K.

Promega UK, Delta House, Chilworth Science Park,

Southampton, SO16 7NS, U.K.

Puritan Medical products company LLC, Guilford,

Maine, USA

Qiagen Ltd, QIAGEN House, Flemming Way, Crawley,

W. Sussex, RH10 9NQ, U.K.

Sanyo E & E Europe BV, Biomedical Division, (U.K.

Office), 9 The office Village, North Road, Loughborough,

Leicestershire, LE11 1QT, U.K.

Sanofi-Aventis, 1 Onslow Street, Guildford, Surrey,

GU1 4XS, U.K.

Sigma-Aldrich Company Ltd, The Old Brick Yard, New

Road, Gillingham, Dorset, SP8 4XT, U.K.

Syngene, Beacon House, Nuffield Road, Cambridge,

CB4 1TF, U.K.

xxi

University of Leicester

VWR

The University of Leicester, University Road, Leicester,

LE1 7RH, U.K.

VWR International LTD, Hunter Boulevard, Magna

Park, Lutterworth, LE17 4NX, U.K.

Introduction 1

1 INTRODUCTION

1.1 Free-living amoebae

Amoebae are both ecologically and medically important, and are named

from the Greek word for change (amoibe), referring to their continual changing

shape as they move and feed.

This ecologically important and diverse group are ubiquitous throughout the

environment. Within soil ecosystems naked amoebae consume large numbers of

bacteria, and are thought to have a key role (Rodriguez-Zaragoza, 1994) comparable

to that of flagellates within aquatic systems (Ekelund and Ronn, 1994). Several

amoebae species are significant within marine ecosystems as both consumers and

producers, and some are known to harbour symbiotic algae (Gast et al., 2009).

There are also free-living amoebae (FLA) that have the potential to live a parasitic

lifestyle and are called amphizoic in recognition of their endozoic existence (Page,

1988). Studies of these parasitic amoebae began in the late 19th

century with the

enteric pathogen Entamoeba histolytica (Lesh, 1975).

Amoebae are a group of unicellular eukaryotic microorganisms that include

Chaos carolinense the giant amoebae, the prototype Amoebe proteus, and parasites

such as E. histolytica, Naegleria fowleri and Acanthamoeba spp. Amoebae are

found throughout the environment from pole to pole with habitats including soil,

sand, seawater and freshwater.

Historically its taxonomy has been unsettled and variable, and this continues

today despite new data from a series of recent genomic sequencing studies (De

Jonckheere, 2004, Gast et al., 1996, Horn et al., 1999, Nassonova et al., 2010,

Schroeder-Diedrich et al., 1998, Smirnov et al., 2007, Stothard et al., 1998,

Weekers et al., 1994). Under the original taxonomic system, the phylum protozoa,

Introduction 2

was divided into four groups with amoebae in sarcodina. However a second system

has been proposed classifying eukaryotes into six supergroups, with amoebae in

Amoebozoa (Adl et al., 2005). All taxonomic classification difficulties are further

compounded because amoebae are a polyphyletic group, having arisen from

different branches of the protozoal evolutionary tree (Schuster and Visvesvara,

2004).

Amoebozoa (Lühe, 1913, emend (Cavalier-Smith, 1998)) although

morphological diverse has several distinguishing features. Pseudopodia (Greek:

false feet) are extended from the cell surface to achieve the distinctive amoeboid

locomotion: they are non-eruptive and morphologically variable. Groups commonly

have subpseudopodia. They can be naked and surrounded by only a plasma

membrane, or testate with a partial single-chambered protective layer called a test.

They are usually uninucleate, rarely binucleate but occasionally multinucleate. Most

groups are at least dimorphic and able form protective cysts when environmental

conditions become adverse.

Pseudopodia are used for both locomotion and feeding: As a chemotaxis

response they detect concentration gradients of nutrients and other substances in

their surrounding environment. The pseudopodia extend and retract with

cytoplasmic streaming (Allen, 1961, Taylor, 1977) which results in the flow of

organelles through the cytoplasm within the cell. Only in the trophic form, can the

amoebae move and feed on surrounding nutrients and microorganisms by

phagocytosis. Pseudopodia surround the food and engulf it, before digestion occurs

within a phagolysosome. The cyst phase is a dormant resting stage, maintained until

the surrounding environment becomes favourable again.

Introduction 3

Several amoebae spp. have been recognised as models for studying cell

mechanisms, one such species is A. castellanii, (ATCC 30010, Neff strain), which

has been used extensively. However a variety of different species have been used to

investigate and research areas including, means of locomotion (Allen et al., 1965,

Klopocka et al., 2009, Pollard, 1981, Sinard et al., 1989), phagocytosis (Obaray and

Coakley, 2001), chemotaxis response (Levchenko and Iglesias, 2002), intracellular

communication (Mogoa et al., 2010, Prusch and Roscoe, 1993) and cell

differentiation (Eichinger et al., 1999). The ability to encyst within a protective

outer layer when conditions become adverse provides protection from desiccation,

starvation, temperature extremes and many chemical disinfectants. Some groups

specifically Naegleria, can also have a flagellate stage.

In some circumstances, amoebae may act as environmental reservoirs, and

be exploited to take up microorganisms, which survive phagocytosis and multiply

within the amoebae (Axelsson-Olsson et al., 2005). Associations such as these may

represent how bacteria adapted and evolved to survive within eukaryotic cells

(Molmeret et al., 2005). Amoebae have even been described as the Trojan horse of

some human diseases (Alsam et al., 2006, Barker and Brown, 1994, Greub and

Raoult, 2004): With some of these microorganisms becoming more virulent whilst

within their amoebae host (Cirillo et al., 1997).

Acanthamoeba can act as a host to Staphylococcus aureus (MRSA) (Huws

et al., 2005), Legionella spp. (Rowbotham, 1998), Clamydia-related symbiont

(Horn and Wagner, 2004), Cryptococccus neoforans (Steenbergen et al., 2001), L.

monocytogenes (Ly and Muller, 1990), Mycobacterium spp. (Adekambi et al.,

2006), Compylobacter jejuni (Axelsson-Olsson et al., 2005), and Escherichia coli

Introduction 4

serotype K-1 (Alsam et al., 2006) and serotype O157 (Barker et al., 1999), and may

act as a vector to transmit these pathogens to susceptible hosts, providing a route of

entry into the human body. The exact mechanisms of bacterial-amoebae interaction

are not understood. As amoebae feed on bacteria, how some pathogenic bacteria

survive within the amoebae, while other non-pathogens are killed, is unclear (Alsam

et al., 2006).

The term endosymbiotic is commonly used to describe microorganisms

within amoebae and their relationship to each other. However studies have shown

microorganisms do not represent true endosymbionts of amoeba, since they can be

both endosymbiotic or lytic, depending on environmental conditions, so the term

‘amoeba-resistant bacteria (ARB)’ has been coined as an alternative (Greub et al.,

2004).

Amoebae have increasingly been recognised as important pathogens, with

the cases of amebiasis increasing in frequency (Martinez, 1997). Infection occurs

largely in immunocompromised rather than otherwise healthy individuals.

Medically important amoebae that bridge the gap between parasitic and free-living

include Acanthamoeba spp., Balamuthia mandrillaris, N. fowleri, and most recently

Sappina pedata (Gelman et al., 2001, Qvarnstrom et al., 2009). Disease

pathophysiology helps to distinguish the infecting agent.

Acanthamoeba and Balamuthia are opportunistic pathogens, which can

cause devastating infections in immunocompromised individuals (either

granulomatous amoebic encephalitis (GAE) or cutaneous infections).

Acanthamoeba can also cause a non-opportunistic ocular infection known as

amoebic (or acanthamoeba) keratitis (AK). Naegleria fowleri causes a fatal non-

opportunistic meningoencephalitis known as primary amoebic meningoencephalitis

Introduction 5

(PAM). Since the discovery of encephalitis caused by S. pedata in an otherwise

healthy young man is so recent, this suggests that there are probably other amoebae

capable of causing fatal infections in humans (Qvarnstrom et al., 2009).

1.2 Acanthamoeba

Acanthamoeba was first observed as a contaminant of trypsinised monkey-

kidney cells in tissue cultures in 1957 (Jahnes et al., 1957), but it was not until the

following year that the pathogenic potential of these amoebae was uncovered during

the development of a polio vaccine (Culbertson et al., 1958). Culbertson and

colleagues used contaminated monkey-kidney cell culture fluids to inoculate

immunosuppressed monkeys and mice, resulting in fatal encephalitis or

encephalomyelitis of the hosts within days. Initially the contaminant was thought to

be an unknown virus. However, upon histological examinations of the monkeys and

mice, and microscopy of the culture fluids the contaminant was identified as

Acanthamoeba. The first reported human infection by Acanthamoeba was described

in 1972 (Jager and Stamm, 1972). It was not until almost a decade later that

Acanthamoeba was characterised as an opportunistic organism, which infects

debilitated or chronically ill patients (Martinez, 1980).

These parasitic FLA are not as well adapted to parasitism as classic parasitic

protozoa such as Plasmodium spp. and Leishmania spp. Not only do the amoebae

invariably kill their host, but also they have not yet evolved to survive within them

for long enough to ensure transmission to a new host. Consequently host-to-host

transmission of these amoebic diseases has not yet occurred.

Introduction 6

Acanthamoeba are incredibly versatile and resilient, as is demonstrated by

their ubiquitous distribution throughout nature and man-made habitats. Within soil

and water samples they are documented as being the most common amoebae if not

most common protozoan (Page, 1988), and they are found almost everywhere there

is water. Their tolerance of marked extremes in temperature, osmolarity, oxygen

availability, water potential and pH allows them to survive in such a wide variety of

locations. As does their ability to encyst when conditions become especially adverse

and survival is threatened.

Such universal distribution means they are not dependent on a host for

transmission and spread. Surprisingly, despite the potential opportunities, infection

of humans by Acanthamoeba is rare, and largely limited to immunocompromised

hosts. Acanthamoeba are responsible for several forms of disease. They can infect

eyes and cause Acanthamoeba Keratitis (AK) (often occurring in healthy

individuals), as well as cutaneous infections, and disseminated infections of the

lungs, the central nervous system (CNS), and the brain, known as Granulomatous

Amoebic Encephalitis (GAE).

So far today there are approximately 24 named species of Acanthamoeba,

but not all appear to be pathogenic (Visvesvara et al., 2007b). The first to be

isolated was A. polyphaga from dust, it was assigned the name Amoeba polyhagus

(Puschkarew, 1913), but was later re-described, and named A. polyphaga (Page,

1967). A. castellanii was discovered as a contaminant of a yeast culture of

Cryptococccus pararoseus (Castellani, 1930 ), which was later named

Hartmannella castellanii (Douglas, 1930). In 1931, the genus Acanthamoeba was

Introduction 7

established (Volkonsky, 1931) and ultimately distinguished from Hartmannella.

Only then was H. castellanii reclassified as A. castellanii (Volkonsky, 1931).

The genus continued to be the subject of controversy, and was revised,

discarded, and finally redefined: But this time the earlier defining characteristics of

pointed spindles at mitosis, general form, locomotion and the appearance of cysts,

were replaced by more definitive features of cyst structure and presence of

acanthapodia (also known as fine finger-like pseudopodia) (Page, 1967).

As taxonomic techniques have become more refined there is now less

movement of species within the grouping system, which has stopped the historical

practice of using the genus names Acanthamoeba and Hartmannella

interchangeably. Hartmannella is a distinctly different amoeba (Page, 1967), which

has never been reliably associated with human pathologies (Cleland et al., 1982,

Culbertson, 1971, Jager and Stamm, 1972).

1.3 Acanthamoeba ecology

Acanthamoeba are dimorphic, and exist as either a cyst or as a trophozoite.

The trophozoites are the larger active feeding form, and ranges in size from 15 µm

to 45 µm. Distinguishing features of trophic amoebae are fine finger-like

pseudopodia, a nucleus containing a distinctive centrally located nucleolus, and a

contractile vacuole found in the cytoplasm integral to maintaining osmotic

equilibrium. The cytoplasm is granular and contains many mitochondria, lysosomes,

ribosomes, and contractile vacuoles. Cysts are smaller than trophozoites and range

from 10 µm to 25 µm. They are non-motile, resistant forms protected by an obvious

inner and outer wall, and are formed in response to a hostile environment. Amoebae

can exist as a cyst for many years if necessary until more favourable conditions

Introduction 8

return. To continue trophic growth, dormant amoebae emerge through a pore or

ostiole, located on the cell wall at the junction of the ectocyst and endocyst

(Martinez, 1997), and covered by a protective flap or operculum.

Acanthamoeba are extremely abundant throughout nature and found

everywhere associated with water, lakes, ponds, oceans, in soil, and even dust

within the air. They have also been found in many man-made environments,

including baths, heating/ventilation/air conditioning systems, cold water storage

tanks, swimming pools, cooling towers of electric and nuclear power plants,

humidifiers, Jacuzzi tubs, dialysis machines, hospital hydrotherapy pools, dental

irrigation units, bottled water, contact lens solutions, bacterial, fungal and

mammalian cell cultures, and medical equipment (Martinez, 1997, Schuster and

Visvesvara, 2004, Visvesvara et al., 2007b). They have even been isolated from ear

discharge, skin lesions, corneal biopsies, cerebrospinal fluid (CSF), pulmonary

secretions, mandibular autografts, brain necropsies, and nasophayngeal mucosa.

Acanthamoeba can inhabit such a wide variety of locations because they are

tolerant of marked extremes in temperature, osmolarity, oxygen availability, water

potential and pH. Additionally they have the ability to encyst when conditions

become especially adverse and survival is threatened. Studies have shown cysts can

survive freeze thaw refreeze cryotherapeutic methods reaching temperatures of

between -50 to -130C (Meisler et al., 1986) although results from a later study

suggest that cyst survival is inversely linked to rate of freezing (Matoba et al.,

1989). Acanthamoeba cysts can also endure periods of up to 8 months at low

temperatures of –10, 4, 10 and 15C (Biddick et al., 1984), pH 2.0, moist heat of

Introduction 9

60C for 60 min (Kilvington, 1991), gamma irradiation (250 K rads) and ultraviolet

radiation (800 mJ/cm2) (Aksozek et al., 2002).

Acanthamoeba have also been shown to remain viable over extended periods

of time. Cysts stored in a state of desiccation for 20 years have hatched and resumed

trophic growth (Sriram et al., 2008).

However some success has been shown in inactivating Acanthamoeba cysts

using solar disinfection (a simulated global solar irradiance of 850 WM-2

) providing

the water containing the organisms reaches a temperature of between 50-55C for 6

or 4 hours respectively (Heaselgrave et al., 2006).

Several environmental factors govern the distribution of Acanthamoeba,

namely organic matter. Bacteria must be readily available for the amoebae to feed

upon and thrive. Consequently Acanthamoeba flourish within biofilm, where food is

in continuous supply. Temperature is also significant, and must be optimal for

strains to thrive; however most can withstand a broad range. Some strains,

particularly clinical isolates, have enhanced growth at warmer temperatures

(>37C). Those isolated from corneal infections generally have an optimal growth

temperature of ~30C (Schuster and Visvesvara, 1998). Pathogenic strains of

Acanthamoeba able to survive within mammals must be able to actively grow at

temperatures of at least 37C. Thermotolerant strains have been isolated from many

biotopes. Including soil samples collected from the island of Tenerife, Canary

Islands, Spain, which showed 90.6% (39 of the 43 isolates) of those collected,

exhibited thermotolerance at 37C (Lorenzo-Morales et al., 2005). While samples

collected from soil in Talbot County, Maryland, USA showed 81.9% (17 of the 21

isolates) displayed thermotolerance between 37-39C (Sawyer, 1989).

Introduction 10

Thermotolerance is not compulsory for species that infect the cornea, as

corneal temperature is around 5C lower than the body between 32-35C,

consequently most species of Acanthamoeba have the potential to become

opportunistic and colonise the corneal surface.

1.4 Acanthamoeba cultivation

As a general rule Acanthamoeba isolates adapt readily to axenic growth, and

as a consequence are an attractive model to use for morphological, biochemical,

nutritional and molecular studies: Often with the type strain A. castellanii Neff

(ATCC 30010) as the most popular choice of species.

Axenic growth for Acanthamoeba is established through a series of steps

based on the ‘walk out’ method (Neff, 1958). In this method amoebae are isolated in

vitro on non-nutrient agar (NNA) plates covered with a monolayer or streak of

gram-negative bacterial food such as E. coli or Enterobacter aerogenes. The

bacteria do not have a food source present and so cannot multiply. Any amoebae

present will crawl as they feed on the bacteria, producing plaque-like clearings in

the bacterial lawn during early growth stage. Once the food source/bacteria has been

exhausted the Acanthamoeba will encyst.

Providing the plates are sealed so they cannot dry out and are stored at 4C,

the cysts will remain viable for long periods of time. Additionally fresh amoebae

cultures can be maintained within the laboratory indefinitely, by transplanting a

small piece of agar containing trophozoites or cysts on to a fresh bacteria-seeded

NNA plate, and placing it isolate-side down. However it must be noted that this

method does not guarantee to establish monocultures of all isolates in a mixed

Introduction 11

sample, as the slower growing amoebae are likely to be outrun by any faster

dividing amoebae.

To establish axenic growth, amoebae are dislodged from the plate surface

using a weak saline solution and are washed by centrifugation removing extraneous

bacteria, before being placed into sterile flasks containing liquid culture medium

and antibiotics. The most effective combination of antibiotics used to inhibit

residual bacteria within this enriched culture medium is penicillin and streptomycin,

or gentamycin. This method works for most isolates but some, in particular clinical

specimens, may require additional nutrients such as foetal calf serum and further

vitamins (Schuster, 2002).

There are several considerations that should be made when designing

experiments using a strain of Acanthamoeba that has been cultured for many

generations. As several studies have demonstrated that traits of strains can change

over long-term culture. It has been shown that strains can develop a reduced

tolerance to temperature, after long-term axenic culture (Pumidonming et al., 2010).

While some exhibit a reduced resistance to therapeutic agents, shown by the

comparison of two cultures of A. polyphaga Ros (one having been in continuous

culture since 1991, and the other cryopreserved since 1991), both tested against a

multipurpose contact lens solution (containing polyhexamethylene biguanide

(PHMB) 1mg/ml) (Hughes et al., 2003).

There may be many other traits that could also have been lost when cultured

for a substantial period of time in an artificial environment. In the case of assessing

the efficacies of contact lens disinfectant solutions against bacteria and fungi a limit

has been set, ensuring strains are passaged no more than five times from the original

Introduction 12

culture, perhaps a similar limit should be taken into consideration when

experimenting with Acanthamoeba.

1.5 Acanthamoeba epidemiology

There are surprisingly few cases of Acanthamoeba infections considering the

ubiquitous nature of the amoebae and the opportunities for contact. However,

unsurprisingly it has been shown that more than 80% of the normal population

possesses anti-Acanthamoeba antibodies (Chappell et al., 2001, Cursons et al.,

1980): Allowing most typically healthy individuals the ability to combat

Acanthamoeba and not develop an infection when exposed to the amoebae. For an

infection to become established, specific predisposing factors must occur.

Breaks in the skin are the most likely mode of infection into the body.

However Acanthamoeba have been isolated from the nasal mucosa of healthy

individuals (Cerva et al., 1973), probably carried there as cysts on air currents or the

wind (Rodriguez-Zaragoza, 1994). So Acanthamoeba may be able to invade through

the upper respiratory tract, again given the right circumstances (Schuster and

Visvesvara, 2004). However the disease incubation period is unknown and weeks to

months may pass following an infection before the symptoms become apparent. As

a consequence of this time delay and the wide distribution of the amoebae within

the environment the precise portal of infection is masked. Once inside the body,

amoebae can spread through the blood to organs and the CNS.

Acanthamoeba infections within the body can result in a host of diseases

including GAE, nasopharyngeal, cutaneous, and disseminated infections, in addition

to AK affecting the eye. GAE is an opportunistic disease, affecting

immunocompromised individuals and therefore has no pattern of seasonality and

Introduction 13

can occur at anytime of the year. However the incidence of GAE has increased with

the HIV/AIDS epidemic, because as a consequence more people are now

immunosuppressed and are therefore susceptible.

Acanthamebiasis primarily occurs in debilitated often chronically ill patients

who are immunocompromised or immunosuppressed often with HIV/AIDS, or who

have undergone organ transplants.

However some cases have been described in individuals who seem not to be

immunocompromised (Singhal et al., 2001). Acanthamebiasis has also been

reported in horses (Kinde et al., 2007), dogs (Dubey et al., 2005, Pearce et al.,

1985), a toucan (from which the etiological agent was isolated, cultured, grown at

44C and therefore shown to be thermotolerant, and genotyped to the T4 group

(Visvesvara et al., 2007a), turkeys, sheep, a kangaroo, reptiles, amphibians, fish

(Dykova and Lom, 2004), invertebrates, pigs, rabbits, pigeons and cattle (Cerva,

1981, Cirillo et al., 1997, Kadlec, 1978, Schuster and Visvesvara, 2004).

Models for encephalitis using the mouse as host have provided information

on virulence, epidemiology and the course of the disease (Mazur et al., 1995). Not

all species, strains and isolates of Acanthamoeba are pathogenic, and those that are,

vary in their virulence. Acanthamoeba virulence is measured by the period of time

from inoculation to the onset of symptoms leading to death, and the number of

animals that have died as a consequence, as well as the dose of inoculants required

to cause encephalitis (Schuster and Visvesvara, 2004). Virulence is also indicated

by the ability of Acanthamoeba to cause cytopathology in tissue cultures (Cursons

et al., 1980, De Jonckheere, 1980, Niszl et al., 1998).

Some strains maintained in axenic conditions for prolonged periods of time,

have been shown to lose their virulence (Mazur et al., 1995) (although this is not

Introduction 14

always the case (Niszl et al., 1998)), their encystment capacity, and change their

susceptibility to drug treatments (Hughes et al., 2003). However most attenuated

characteristics including virulence can be restored if passaged in series through

human Hep-2 cell monolayer (Kohsler et al., 2009) or in vivo via intranasal

inoculation into mice and ultimately the brain (Xuan et al., 2009).

Acanthamoeba is the etiological agent for AK. The disease can occur in

otherwise healthy individuals who almost always are contact lens wearers.

Observations made studying rat and mouse models have shown mode of infection

usually occurs following a trauma to the eye and then wearing contact lenses (Ren

and Wu, 2010). With poor contact lens (and lens case) hygiene being a major

contributing factor to acquiring the disease. Infection does not automatically occur

in both eyes. However, there has been one case of fatal GAE with associated uveitis

and pharyngitis recorded in the literature, but never a case of GAE having

developed from AK (Jones et al., 1975, Visvesvara, 2010).

The precise incidence of AK is unknown, as the disease is not notifiable,

however it has undoubtedly been on the increase as more people wear contact lenses

today than previously. There was a recent outbreak of AK detected in the USA in

2006. Initially the cases were identified within the Chicago area by Illinois

Department of Public Health, and brought to the attention of the Centers for Disease

Control and Prevention (CDC). Who performed several surveys, one of which was a

retrospective survey of ophthalmology centres across the USA. Their results

showed an increase in the number of culture-confirmed cases during 2004-2006

compared with 1999-2003, in ten centres over nine states. Following a national

outbreak investigation, analysis indicated the odds of ever having used the

Introduction 15

multipurpose contact lens solution, Complete MoisturePlus (AMOCMP)

manufactured by Advanced Medical Optics (now known as Abbotts Medical

Optics) were 20 times greater for AK patients than for controls. Following

communication with the food and drug administration (FDA, AMO voluntarily

recalled AMOCMP worldwide (Verani et al., 2009).

1.6 Granulomatous amoebic encephalitis (GAE)

Acanthamoeba are one of the causative agents of granulomatous amoebic

encephalitis also known as GAE. It is a rare and often fatal infection, found most

commonly in immune compromised or severely debilitated individuals. GAE was

first observed in 1972 (Jager and Stamm, 1972) with less than 200 cases caused by

Acanthamoeba documented within the literature (Schuster and Visvesvara, 2004).

This however is likely to be a false representation, as these infections are difficult to

diagnose and distinguish from bacterial and other microbial infections even in the

first world.

Headaches, slight fever, behavioural abnormalities, personality changes, stiff

neck, nausea, hemiparesis, seizures, cranial nerve palsies, and typical signs of

localised encephalopathy are all symptoms of the chronic progressive disease (da

Rocha-Azevedo et al., 2009, Marciano-Cabral, 2003, Schuster and Visvesvara,

2004, Walochnik et al., 2008). As the clinical signs are not specific, the disease is

often misdiagnosed. Differential diagnosis includes bacterial meningitis, viral

encephalitis, neurocyticerosis, and brain tumors (da Rocha-Azevedo et al., 2009,

Matson et al., 1988, Ofori-Kwakye et al., 1986).

Introduction 16

Diagnosis if made, is almost always post-mortem following brain tissue

biopsies, and indirect immunofluorescence (IIF) staining of tissue sections (Bloch

and Schuster, 2005, Schuster et al., 2006b, Schuster and Visvesvara, 2004).

Medical image scanning to view the brain, such as computerised

tomography (CT) and magnetic resonance imaging (MRI) can be used to visualise

the lesions caused by Acanthamoeba, but these lesions are not specific enough to

the disease to base a diagnosis on (McKellar et al., 2006).

A diagnosis of GAE has been made pre-mortem, and was achieved as a

result of positive Acanthamoeba-specific PCR of several biopsy tissue and fluid

specimens, including cerebrospinal fluid (CSF), bronchoalveolar lavage specimens

(BAL), skin, lung, and brain tissue (from the main lesion): All of which had been

Acanthamoeba culture negative when tested (Walochnik et al., 2008).

PCR is a proven invaluable diagnostic tool, used to detect Acanthamoeba

infections. A reliable primer pair designed and developed from the complete DNA

gene sequence of the 18S ribosomal gene (18S rDNA: Rns), known as JDP1 and

JDP2, have the ability to detect all known Acanthamoeba subgroups successfully,

including those from the environment, AK and GAE patients (Stothard et al., 1998).

These primers within a PCR assay produce a specific amplimer of around 500 bp,

and from this sequence variation has allowed the development of a highly sensitive

T-group typing system. The system has even been used to determine an

epidemiological association between a keratitis-causing strain of Acanthamoeba, the

patient, their contact lens storage case and their domestic water supply (Ledee et al.,

1996). On going studies have identified T-groups 1-15, with the majority of GAE

and AK causing amoebae in the T4 subgroup (Gast, 2001, Hewett, 2003, Horn et

al., 1999, Schroeder et al., 2001, Stothard et al., 1998).

Introduction 17

T-group typing is not the only molecular diagnostic technique for detecting

Acanthamoeba. Comparison of sequence variation of the mitochondrial 16S rRNA

genes (Ledee et al., 2003), Restriction fragment length polymorphisms (RFLP)

nuclear and mitochondrial rRNA (Kilvington et al., 1991, Kilvington et al., 2004),

and fluorescent probes to hybridise with Acanthamoeba DNA (Stothard et al.,

1999). Real time PCR assays have also been highly successful at determining

Acanthamoeba infections; they are rapid and particularly sensitive, with the ability

to be effective even with a low concentration of template DNA (less than 10 cells)

(Qvarnstrom et al., 2006, Riviere et al., 2006).

In addition to the molecular diagnostic techniques available to identify

Acanthamoeba infections, there are also many molecular methods. Although rare,

amoebae can be identified in wet preparations of CSF samples or those that have

been giemsa or H & E stained (Marciano-Cabral, 2003, Martinez, 1997, Seijo

Martinez et al., 2000, Sharma et al., 1993). Identification of amoebae in both

trophozoite and cyst stages can also be made from biopsied patient samples, which

have been formalin-fixed or paraffin-embedded (da Rocha-Azevedo et al., 2009),

using techniques including immunohistochemistry and fluorescent microscopy.

Rabbit generated, anti-Acanthamoeba antibodies are incubated with patient samples,

followed by a secondary incubation with anti-rabbit IgG associated with a

fluorescent marker such as FITC, and observed under fluorescent microscopy

(Culbertson and Harper, 1984). More recently species-specific monoclonal

antibodies for A. castellanii, A. polyphaga, A. lenticulata, and A. culbertsoni have

been developed (Guarner et al., 2007).

Introduction 18

Typically GAE infections occur in the CNS tissue, but may involve the

lungs (Khan, 2006). The incubation period is unknown, and weeks to months may

elapse before the onset of the disease becomes apparent. This time delay obscures

the precise route of entry of the amoebae into the body, but the most likely portal is

the skin, olfactory neuroepithelium (Visvesvara et al., 2007b, Walochnik et al.,

2008) and in specific circumstances orally (Thamprasert et al., 1993). Once inside

the body, the amoebae most likely gain access into the CNS and lungs by

hematogenous dissemination, or by passing directly through the neuroepithelium

(Marciano-Cabral, 2003).

Acanthamoeba can also cause lesions in the skin, but these are more often

reported in association with HIV positive patients (da Rocha-Azevedo et al., 2009,

Torno et al., 2000). Cutaneous acanthamebiasis presents with multiple hard

erythematous nodules, papules, or ulcers, across the surface of the body (May et al.,

1992). If they occur simultaneously with CNS symptoms their presence is often

indicative of an infection by Acanthamoeba (da Rocha-Azevedo et al., 2009, Khan,

2006).

Pathogenesis of GAE is extremely complex and not yet fully understood,

and as a result, prognosis is extremely poor. As the majority of people who contact

GAE are often immune compromised, debilitated and/or chronically ill, these

predisposing factors are further compounded by the difficulty associated with

diagnosis. Treatment of Acanthamoeba infections is further hampered by the

amoebae ability to encyst when environmental conditions are detrimental (such as

in the presence of treatment drugs) and remain so until conditions become more

favourable.

Introduction 19

However several patients have been diagnosed early enough and have

subsequently been successful treated (Lackner et al., 2010, Seijo Martinez et al.,

2000, Walia et al., 2007, Walochnik et al., 2008). Each of these patients has been

treated with a different antimicrobial, as often one drug treatment used successfully

in one patient is ineffective in another. Several factors are influential to the

outcome of the disease, these include: how early drug treatments are initiated; host

immune status; infective dose of the amoebae; antimicrobial sensitivity of the strain

and its virulence (Schuster and Visvesvara, 2004).

Drug treatments used clinically include amphotericin B, azithromycin,

fluconazole, 5-fluorocytosine (flucytosine), pentamidine isethionate, meropenem,

linezolid, moxifloxacin, miltefosine, amikacin, voriconazole and sulfadiazine

treated (Lackner et al., 2010, Schuster and Visvesvara, 2004, Seijo Martinez et al.,

2000, Walia et al., 2007, Walochnik et al., 2008).

1.7 Acanthamoeba keratitis (AK)

Acanthamoeba keratitis (AK) is the sight threatening, acute, progressive and

extremely painful ulceration of the cornea caused by species of Acanthamoeba. The

infection almost always occurs in immune competent individuals who are contact

lens wearers, or who have minor corneal abrasions.

AK was first recognised in 1973 in a rancher from South Texas, USA, who

had a history of trauma in one eye (Jones et al., 1975). Since then, year on year the

numbers of cases reported within the literature have continued to rise, however the

infection is not classified as a notifiable disease. Contact lenses are a risk factor

associated with AK, primarily in the users of soft contact lenses (Stehr-Green et al.,

1989). Incidence levels for the USA are estimated at an annual incidence of 1-2

Introduction 20

cases per million contact lens wearer (Schaumberg et al., 1998). Mean while the

estimates for the U.K. are higher at 1 in 30,000 contact lens users (Seal, 2003). In

one decade between 1990 and 2000, 180 cases from the U.K. were detailed in the

literature (Radford et al., 1995, Radford et al., 1998, Radford et al., 2002) with most

patients referred for treatment to Moorfields Eye Hospital, London, U.K.. Several

factors are likely to be linked to the observed increase in AK cases, namely the

better understanding of the infection we have now, which has led to an increase in

the number of correct diagnoses made, and the years on year increase in number of

people wearing contact lens.

Typically AK occurs in one eye only, but bilateral AK has been described,

often developing as a complication of the initial infection (Lee and Gotay, 2010,

Wilhelmus et al., 2008). Symptoms of AK are not specific initially, and include

severe eye pain, eye redness, photophobia, blurred vision, and a sensation of

something in the eye, along with excessive tearing. Clinical symptoms develop to

include corneal inflammation leading to the formation of a ring-like stromal

infiltrate, corneal oedema, and erosion of the corneal epithelial cells.

AK suffers experience disproportionate eye pain, which thought to be linked

to radial keratoneuritis and the trophozoites found along the corneal nerves, leading

to thickening and distortion of these nerves (Yoo et al., 2004). Later stages of the

infection can result in epithelial denudation, stromal necrosis (da Rocha-Azevedo et

al., 2009), nerve oedema and retinal detachment, and if miss-diagnosed or a delay in

treatment occurs, the infection will almost certainly lead to blindness as the necrotic

region spreads inwards (Niederkorn et al., 1999).

Introduction 21

A waxing and waning of the clinical course is usually a good diagnostic

indicator of AK. This trend is caused because Acanthamoeba are often partly

susceptible to non-protozoan treatments. An additional potential complication of

diagnosis and therefore management is a secondary infection due to bacteria.

Differential diagnosis of AK should be considered, as keratitis caused by Herpes

simplex has dendriform-appearing lesions on the cornea (Martinez 1997), while

Pseudomonas aeruginosa, has a similar type of infiltration (Clarke and Niederkorn,

2006).

AK can be diagnosed using the same techniques as for GAE, however cyst

and trophozoite stages can also be detected and identified in ocular samples, either

from biopsies or less invasive corneal scrapes, however considerable expertise is a

necessity. The samples are smeared onto glass slides and visualised with bright field

microscopy. Or wet mount preparations can be made using 10% KOH (Bharathi et

al., 2006). Wet mounts stained with Giemsa or with H & E can be used to readily

detect cysts, while trophozoites are harder to distinguish, as they resemble

inflammatory cells (Bharathi et al., 2006, da Rocha-Azevedo et al., 2009). Again, as

with GAE, IIF, monoclonal antibodies and fluorescent microscopy can be used to

detect Acanthamoeba in corneal samples, contact lenses and lens cases (Inoue et al.,

1999, Leher et al., 1999, Mietz and Font, 1997). There are several fluorescent

stains effective against Acanthamoeba including calcofluor white, fungiflora,

Acridine orange, Periodic acid-Schiff (PAS), and Gomori methamine silver (GMS)

(Hahn et al., 1998, Inoue et al., 1999, Vemuganti et al., 2000, Wilhelmus et al.,

2008). Confocal microscopy has also been shown to be useful in detecting AK

Introduction 22

(Vaddavalli et al., 2011), but specialised equipment and experienced observers are

required.

Molecular techniques highlighted for diagnosing GAE infections as effective

with AK. Such techniques have made it possible to trace the source of AK

infections back to taps in the patient’s home (Booton et al., 2002, Kilvington et al.,

2004, Ledee et al., 1996).

Acanthamoeba are able to infect a cornea and establish an infection, when

specific circumstances arise. If there is corneal trauma in association with

contaminated water, or the wearing of contact lenses maintained with a poor

hygiene routine. AK infections principally occur in individuals, who wear soft

contact lenses, although infections have been documented in some wearing hard

contact lenses as well (Moore et al., 1987, Srinivasan et al., 1993). Acanthamoeba

cyst more readily attach to soft contact lenses which have a higher water content

than rigid hard lenses (Sharma et al., 1995).

Epidemic factors also linked to AK are swimming in contact lenses (Radford

et al., 2002), a failure to maintain proper lens care i.e. using unsterile contact lens

solutions, inadequate disinfection of the lenses, not soaking the lenses for the full

amount of their recommended period, storing the lenses in a dirty case and/or

washing the lenses in homemade saline solutions (Martinez and Visvesvara, 1997,

Radford et al., 2002).

As a result of wearing contact lenses (hard or soft), minor abrasion of the

corneal epithelium can occur, allowing the infection to take hold (Visvesvara et al.,

2007b). Once Acanthamoeba trophozoites have adhered to the surface of the corneal

epithelial cells, mediated by mannose-bonding protein (MBP) expressed on the

Introduction 23

surface of the amoeba, damage is inflicted by phagocytosis of the host cells (Clarke

and Niederkorn, 2006). In vitro studies with Acanthamoeba and rat B103

neuroblastoma cells, viewed by scanning and transmission electron microscopy,

have shown phagocytosis. Following contact between the two cells types,

membrane blebbing of the nerve cells occurred, leading to either lysis of the nerve

cell, or ingestion of the nerve cell via amebastome-like (food cups) and channelling

the ingested cell into intracytoplasmic food vacuoles (Pettit et al., 1996).

To ensure a good prognosis, early diagnosis and appropriate treatment of

AK is vital. The first successful treatment of AK was reported in 1985, using an

antimicrobial agent belonging to the aromatic diamidine group of compounds,

known as propamidine isethionate and available as Brolene 0.1% w/v (Sanofi-

Aventis, Guilford, U.K.). This used in combination with Neosporin® (Johnson &

Johnson, New Brunswick, USA), and was Europe’s first-line medical treatment

until the mid 1990’s. When resistance to propamidine isethionate was detected in

some strains of Acanthamoeba and studies showed it to be only weakly cysticidal

(Perrine et al., 1995). The search has continued for other superior homologues and

alternative drug treatments such as the bis-biguanides, alexidine and chlorhexidine.

Alexidine has a cytopathic affect on Acanthamoeba, not unlike chlorhexidine, with

10 mg/mL being affective against trophozoites and 100 mg/mL against cysts. In

vivo experiments carried out using Chinese hamster corneas showed alexidine to be

less toxic to corneal epithelial cells than chlorhexidine at 100 mg/mL (Alizadeh et

al., 2009). More recently in vitro studies have shown Acanthamoeba to be

susceptible to PHMB (Hughes et all 2003), showing consistent cysticidal activity.

Introduction 24

Now the recommended treatment regimes for AK are chlorhexidine with

propamidine, or its polymeric equivalent PHMB originally with propamidine but

now combined with hexamidine isethionate (Hexomedine, Sanofi-Aventis) (Seal,

2003). The biguanides are a class of cationic disinfectants, and the topical use of

them is recommended as the only effective therapy for resistant encysted forms of

Acanthamoeba in vitro, and likely in vivo (Dart et al 2009). If the treatment is

administered promptly enough a response to AK is expected within one week, with

total eradication of the active infection to occur within eight weeks. A stark contrast

to the four months of treatment needed if using Brolene and neomycin (Seal, 2003).

If the infection has not responded to topical biguanide, the next treatment to

attempt is the use of steroids, however this is controversial (Dart et al., 2009). As a

last resort in extreme cases that have shown no signs of improvement therapeutic

keratoplasty is then recommend (Dart et al 2009).

1.8 Taxonomy and Classification

So far today there are approximately 24 named species of Acanthamoeba

based on morphological characteristics (Schuster and Visvesvara, 2004, Visvesvara

et al., 2007b). Several of which do have pathogenic potential, and include A.

polyphaga, A. rhysodes, A. quina, A. griffini, A. lugdunensis, A. castellanii, A.

culbertsoni, A. healyi and A. hatchetti (Ledee et al., 1996, Schaumberg et al., 1998,

Schuster and Visvesvara, 2004, Yu et al., 2004).

The genus Acanthamoeba, is divided into three morphological categories (I,

II, and III) based on distinguishing features of the cyst, including diameter, and

shape of both endocyst and ectocyst walls (Pussard and Pons, 1977). Today the

most comprehensive dichotomous key for morphological taxonomy is based upon

Introduction 25

the three group system, and is nearly 20 years old (Page, 1988). Acanthamoeba

cysts are double walled, with an inner layer or endocyst composed of cellulose, and

an outer made up of lipids and proteins known as the ectocyst (Blanton and

Villemez, 1978). Group I amoebae have large cysts with a diameter of 18 µm or

more, with stellate endocysts and smooth or wrinkled ectocysts (Table 1). Amoebae

in group-II are the most common. This group is also the largest and includes the

majority of the potentially pathogenic species (Table 1). Their cyst diameter is 18

µm or less, with polyhedric, globular, ovoid, or stellate endocysts and wavy

ectocysts. Group III amoebae have cysts of less than 19 µm with ovoid or globular

endocysts and wavy or smooth ectocysts (Table 1). The reliability of using cyst

morphology as a taxonomic characteristic came into question once it was observed

that ionic concentration of the growth media can alter cyst morphology (Sawyer,

1971).

Immunological, biochemical and physiological methods have been used in the

past to attempt to type and identify Acanthamoeba species. Techniques used include

western blotting and immunofluorescence, however they have been inconclusive at

distinguishing between Acanthamoeba species.

More recently a system has been proposed to classify Acanthamoeba based

on 18S rDNA (Rns) gene phylogeny (Gast et al., 1996, Schroeder-Diedrich et al.,

1998). 18S ribosomal RNA is the structural RNA for the small component of the

eukaryotic cytoplasmic ribosome: Which has a slow evolutionary rate making it an

ideal candidate to study and base reconstruction of ancestral divergences on.

Acanthamoeba are grouped based on sequence variation into evolutionary clades.

Introduction 26

So far 15 clades have been recognised, each containing species or complexes of

closely related species: T1-T12 (Gast et al., 1996, Schroeder-Diedrich et al., 1998),

T13 (Horn et al., 1999), T14 (Gast, 2001), and T15 (Hewett, 2003). The group T4

contains the majority of species with pathogenic potential (Stothard et al., 1998).

Table 1, shows examples of morphologically designated Acanthamoeba, and how

they are grouped according to T-group genotyping.

Table 1. Morphologically designated Acanthamoeba species assigned to T-groups

based on phylogenetic analysis of 18s sequence variations.

T

group

Species Strain ID Reference Morphological

group

T1 A. castellanii ATCC 50494 (Dudley et al.,

2005)

II

T2 A. palestinensis ATCC 30870 (Gast et al.,

1996)

III

T3 A. griffinii ATCC 30731 (Gast et al.,

1996)

II

T4 A. rhysodes

A. polyphaga

A. castellanii

A. lugdenensis

ATCC 50368

ATCC 30971

ATCC 50374

L3a

(Gast et al.,

1996)

(Stothard et al.,

1998)

(Gast et al.,

1996)

(Kong, 2009)

II

II

II

II

Introduction 27

T5 A. lenticulata ATCC 30841 (Stothard et al.,

1998)

III

T6 A. palestinensis ATCC 50708 (Stothard et al.,

1998)

III

T7 A. astonyxis ATCC 30137 (Stothard et al.,

1998)

I

T8 A. tubiashi ATCC 30867 (Stothard et al.,

1998)

III

T9 A. comandoni ATCC 30135 (Stothard et al.,

1998)

I

T10 A. culbertsoni ATCC 30171 (Stothard et al.,

1998)

III

T11 A. hatchetti

A. stevensoni

Sawyer:NMFS

RB:F1

(Stothard et al.,

1998)

(Kong, 2009)

III

II

T12 A. healyi CDC1283:V013 (Stothard et al.,

1998)

III

T13 Acanthamoeba

sp. *

- (Horn et al.,

1999)

-

T14 Acanthamoeba

sp. *

- (Horn et al.,

1999)

-

T15 A. jacobsi ATCC 30732 (Hewett, 2003) III

*: Without morphological designation; -: Not assigned.

Introduction 28

Still despite the recent advances in identification and typing of

Acanthamoeba and all the information available in the literature there are still many

uncertainties, confusions and complications to unravel. Which is unsurprising,

considering the potential margin of error through misidentification of isolates,

possible mislabelling of culture tubes, and/or cross-contamination of cultures. To

highlight just one example, a group have analysed mitochondrial DNA restriction

fragment length polymorphisms (mt RFLP), as well as sequences of both the

nuclear 18S rDNA and mitochondrial 16S rDNA, of four morphological group II

Acanthamoeba. From their results they have proposed that A. divionensis, A.

paradivionensis and A. mauritaniensis all be regarded as synonyms for A. rhysodes

(Liu et al., 2005).

1.9 Molecular biology of Acanthamoeba

There has been a lack of interest in studying the molecular biology of

Acanthamoeba, but with the advent of recombinant DNA techniques, genomes of

the organisms have now become more assessable for studies. The recent outbreaks

of AK, and the realisation that Acanthamoeba can harbour pathogenic bacteria, and

therefore serve as a vector in human infections, have all boosted the current interest

in Acanthamoeba.

Acanthamoeba replicate by mitosis (Jantzen et al., 1990), with the nuclear

membrane disappearing during division (Ma et al., 1990). As yet there has been no

evidence of sexual reproduction occurring in Acanthamoeba (Yin and Henney Jnr,

1997). Evidence suggests that nuclear chromosomes are numerous and small with

some ranging from 200 Kb to larger than 2 Mb (Byers, 1986, Byers et al., 1990,

Rimm et al., 1988). Acanthamoeba trophozoites normally possess one nucleus,

Introduction 29

which is approximately 16.6% of the size of the trophozoite (Khan, 2006). The

nucleus does not contain any membrane bound sub components, but does contain a

large centrally located nucleolus, which is also a prominent morphological

characteristic.

The genome of A. castellanii Neff has been estimated at 33 Mb, and made

up of approximately 60% GC (Byers, 1986). With a DNA content most commonly

estimated at 1.28 pg/amoeba (Adam et al., 1969). Nuclear DNA content varies

throughout culture growth, and decreases by approximately 50% (in unagitated

cells) during the transition from log to post-log phase, and is most likely linked with

preparations of the cell to encyst (Byers et al., 1969).

Mitochondrial DNA content has also been shown to vary throughout culture,

with measurements of log phase A. castellanii Neff containing an average of 0.15

pg/amoeba, and reducing to 0.01-0.02 pg/amoeba during encystment (Byers, 1986).

In recent years the mitochondrial genome of Acanthamoeba has been fully

sequenced, as part of collaboration to explore mitochondrial genome organisation

and evolution within protists. Acanthamoeba mitochondrial genome is a circular

molecule consisting of 41.6 Kb, with an AT content of 70.6%. Which contains the

genes that code for both large and small subunit rDNA, 16 tRNA, 8 open reading

frames (ORF) of undetermined function and 33 proteins. All of these are found in

the same transcriptional orientation and make up 93.2% of the total sequence

(Burger et al., 1995). A peculiar feature of the mitochondrial genome of A.

castellanii is the presence of a single continuous ORF (cox1/2) encoding subunits 1

and 2 of cytochrome oxidase (COX1 and COX2) (Burger et al., 1995).

Phylogenetic trees based on sequences of nuclear rDNA, place A. castellanii

on a branch with, or near (as an out group to) green algae and land plants. But as

Introduction 30

small subunit SSU rDNA databases have expanded, Acanthamoeba has moved

away from the algal and plant clade, to a sister branch of animals and fungi

(Wainright et al., 1993) or outside of the multicellular lineages altogether (Cavalier-

Smith, 1993). However one group has discovered evidence of horizontal gene

transfer between algal chloroplasts and Acanthamoeba mitochondria (Lonergan and

Gray 1994). By sequencing from the mitochondria of A. castellanii, a 7,778 bp

region containing single-copy large subunit (LSU) and small (SSU) rRNA genes,

they identified three group I introns within the LSU rRNA gene (rnl). The introns

are placed within highly conserved regions and each possesses a freestanding open

reading frame (ORF). One of the introns was found to be identical to that of the

single group I intron in the chloroplast rnl of green algae Chlamydomonas

reinhardtii, and structurally homologous within the core region and the ORF they

encode. Suggesting intron movement has occurred between mitochondria and

chloroplasts, either intracellularly in a photosynthetic, remote common ancestor of

A. castellanii and C. reinhardtii or, more recently as a result of an intercellular

exchange of genetic material (Lonergan and Gray 1994).

With the continued search for new data, this theory has developed as a result

of the complete sequencing of the Acanthamoeba mitochondrial genome. The

overall size of A. castellanii’s mt DNA, its gene content and organisation most

closely resembles that of the chlorophycean algae Prototheca wickerhamii, than C.

reinhardtii. Comparison of the mitochondria from all three organisms shows A.

castellanii and P. wickerhamii to have almost identical respiratory and ribosomal

protein genes, while C. reinhardtii does not encode any ribosomal proteins and

lacks several standard respiratory genes. The authors argue that the results can be

interpreted in two possible scenarios, either C. reinhardtii does not share a common

Introduction 31

ancestry with land plants and A. castellanii, or more likely, that they do all share a

common ancestor, and C. reinhardtii has diverged radically and lost many of its

former gene content (Burger et al., 1995).

Ribosomes and their associated sequences have been extensively studied

throughout biology as a means to better understand evolutionary origins. As a

consequence the ribosomal biology of Acanthamoeba is one of the better-

understood areas of its molecular biology.

One of the most studied and frequently used genes throughout eukaryotic

biology is a component of the small ribosomal subunit 40S, known as the small

subunit (SSU) 18S rRNA. SSU 18S is the structural RNA for the small component

of the cytoplasmic ribosomes, and is therefore integral to protein synthesis in all

living cells. Molecular analysis using 18S data to understand evolutionary

divergences has become extremely popular. Several factors make SSU 18S such a

prime target for molecular analysis, both its slow evolutionary rate and repetition

throughout the genome provide ample template for PCR. Additionally, the gene is

usually flanked with highly conserved regions, therefore making it relatively easy to

locate at the outset of the studies, and so readily accessible.

Amoebae rDNA coding sequences are arranged as is typical for eukaryotic

ribosomal gene repeat units; they contain one set of 5' 18S, 5.8S and 28S 3' genes.

Between these genes and neighbouring sets are spacer regions or internally

transcribed spacers (ITS). Acanthamoeba rRNA repeats unit is 12 Kb, containing

approximately 600 copies (Byers et al., 1990), while the coding sequence for the

18S rRNA gene is 2,303 bp long.

Introduction 32

Acanthamoeba nuclear SSU 18S rRNA genes (18S rDNA; Rns) have been

fairly extensively studied and are at present receiving much attention, in the hope of

understanding why some strains appear pathogenic and others do not. Group I

introns within the rRNA sequence have been found in A. griffinii and A. lenticulata

(Gast et al., 1994, Schroeder-Diedrich et al., 1998), increasing the size of their

nuclear rRNA genes to approximately 2,800 bp, compared with 2,300 bp. Sequence

analysis of the Rns has allowed the development of a classification scheme,

allowing Acanthamoeba to be typed into one of a possible 15 Rns genotypes known

as T-groups (Booton et al., 2005, Gast et al., 1994, Hewett, 2003, Stothard et al.,

1998).

Many amino acid sequences have been studied and published, these include

complete and/or partial, mRNA, RNA and genomic sequences. The literature and

GenBank are dominated by 18S sequences from many Acanthamoeba sp., but other

sequences available include: actin I (Nellen and Gallwitz, 1982); myosin heavy

chains I (Brzeska et al., 1999); 26S (Lai and Henney, 1993); 5S (Zwick et al.,

1991); lactate dehydrogenase-like (Watkins and Gray, 2006); mannose-binding

protein (Garate et al., 2004); polyubiquitin (Hu and Henney, 1997) and profilin I

and II (Pollard and Rimm, 1991).

Although slow to start, the field of Acanthamoeba molecular biology has

moved forwards rapidly. Evidence has recently been presented to include a new

genotype of 18S, T16 (Corsaro and Venditti, 2010). While data also suggests

analysis of microsatellites found within the ITS located between 18S and 5,8S,

could be a potential candidate to further distinguish within the clades, especially T4

(Kohsler et al., 2006).

Introduction 33

1.10 Aims

The aims of this thesis will be to study the molecular biology of

Acanthamoeba with the aim of developing a genotyping system to better resolve the

Acanthamoeba genus. An improved genotyping system could prove invaluable for

epidemiological fingerprinting and species resolution in the currently muddled

group. The new system will be put to the test in a case study of an outbreak of GAE

within a Swedish hospital. The following chapters will first introduce the subject

being investigated, and then outline the aims, methods, and any results found,

before a discussion is presented.

DNA typing of Acanthamoeba sp. 34

2 DNA TYPING OF ACANTHAMOEBA SP.

2.1 Introduction

The classic morphological typing system divides the Acanthamoeba genus

into three groups based on morphological characteristics of the cysts including size

and shape (Pussard and Pons, 1977): Group I species, have relatively large cysts

with a diameter of at least 18 µm, with distinctly stellate endocysts and smooth or

slightly wrinkled ectocysts. Group II Acanthamoeba have a diameter of 18 µm or

less, with polyhedric, globular, ovoid, or stellate endocyst and wavy ectocyst. While

those classified to group III, have cyst diameters of less than 19 µm with ovoid or

globular endocysts, and wavy or smooth ectocysts.

The reliability of using cyst morphology as a taxonomic characteristic came

into question with the observation that variations in the ionic concentration of the

growth media can alter cyst morphology (Sawyer, 1971, Stratford and Griffiths,

1978). As a consequence the morphological classification scheme has been

challenged by the use of isoenzyme electrophoretic patterns to study intragenetic

relationships: Results have been mixed, with some good correlations between

isoenzyme patterns and morphological groups (Moura et al., 1992), and those that

contradicted their morphological group designations’ (Daggett et al., 1985, De

Jonckheere, 1983).

Isolation, cultivation and subsequent microscopy were once the main

method of Acanthamoeba identification. However due to the ambiguities of the

morphological system, these techniques have now been superseded: Although they

remain a vital component, as both a useful method of diagnosis to genus level, and

to cultivate sufficient numbers of cells for downstream assays. Attempts have been

made to develop molecular diagnostic techniques for studying Acanthamoeba and


its associated diseases: These include techniques such as PCR amplification of

nuclear SSU 18S (Gast et al., 1996, Gunderson and Sogin, 1986), mitochondrial

cytochrome oxidase subunit 1 and 2 (cox1/2) gene (Kilvington et al., 2004),

mitochondrial 16S rRNA genes (Chung et al., 1998), as well as PCR-restriction

fragment length polymorphism (PCR-RFLP) analysis of nuclear and/or

mitochondrial rDNA (Yu et al., 1999) and fluorescent probes to hybridise with

Acanthamoeba DNA (Stothard et al., 1999). Ultimately molecular developments

have resulted in a phylogenetic typing system, which is both genus and subgenus-

specific, allowing Acanthamoeba to be classified based on 18S sequence genotype:

This system can assign isolates representing all three morphological groups to one

of a possible 15 T-groups (Booton et al., 2005, Gast et al., 1994, Hewett, 2003,

Stothard et al., 1998), based on sequence similarities of SSU 18S rDNA. The T-

group system has proven to be a sensitive tool, with the ability to detect very small

numbers of trophic amoebae in samples (Schroeder et al., 2001). Eukaryotes have

many copies of rRNA genes found repeatedly throughout the genome, providing an

excess of template available for PCR.

Although the 18S sequence typing system for Acanthamoeba is highly

sensitive and useful, it is not without its own specific complications. Such as

multiple alleles, where some isolates have been found to have three alleles of 18S

(Ledee et al., 1998). A second complication is the presence of Group 1 introns in

some species, such as A. griffinii (S-7; ATCC 30731) and A. lenticulata (PD2S;

ATCC 30841) (Gast, 2001, Gast et al., 1994), which distorts interstrain

relationships. The 18S T-group genotyping system is also not particularly robust

and clumps together both disease causing strains, into single clades with non-

pathogenic environmental strains (Booton et al., 2005). Of the 15 recognised


groups, T4 is by far the largest with little resolution provided between the isolates.

The development of a second system has been attempted to better resolve T4 using

mt 16S rRNA sequences, but was unsuccessful (Ledee et al., 2003).

For a second system to be developed successfully, a gene suitable for

phylogenetic comparisons needs a high level of diversity. A key enzyme in aerobic

metabolism is mitochondrial cytochrome oxidase: Which is a component of the

respiratory chain, and involved with transfer of electrons from cytochrome c to

oxygen. Unusually within Acanthamoeba this gene named cox1/2, encodes for both

subunits of cytochrome oxidase and specified by a single continuous ORF (Burger

et al., 1995). Cytochrome oxidase genes have already been used for phylogenetic

typing systems in a variety of different organisms, including crayfish (Yue et al.,

2008), Echinococcus granulosus (Villalobos et al., 2007) and Trypanosoma cruzi

(Burgos et al., 2008).

The mitochondrial gene cox1/2 is a suitable contender as an alternative

target for sequence analysis typing, to be used as a target for rapid and sensitive

diagnosis of AK. Mitochondrial DNA has a large copy number in Acanthamoeba,

approximately 3,300 for a log phase amoebae of the Neff strain (Byers, 1986), and

analysis has shown it to have a large degree of nucleotide variation. Phylogenetic

analysis using cox1/2 has already been used to differentiate eight patient isolates, as

well as for six of them, matching their sequence homology with their respective tap

water isolates (Kilvington et al., 2004).

Recent studies have shown multiple housekeeping genes can be

concatenated to form a single super-gene alignment for building more robust

phylogenetic trees, which provide better discrimination power (Devulder et al.,

2005, Gadagkar et al., 2005, Kurtzman and Robnett, 2007). The use of the


concatenated gene approach within Acanthamoeba, using the well documented SSU

18S rDNA gene from the T-group system, in combination with another

housekeeping gene, such as mt cox1/2 could yield more accurate trees with greater

powers of discrimination and ultimately provide the potential to phylogenetically

classify the subgenus and improve the diagnosis of AK.

2.1.1 Aims

To investigate the phylogeny and relatedness of multiple Acanthamoeba

species including both environmental and disease causing strains, by amplifying and

analysing two distinct genes, the housekeeping gene mt cytochrome oxidase subunit

1 and 2 (cox1/2) gene, and the well documented SSU 18S rDNA (Rns) gene,

making comparisons between their genotypes and morphological classifications.

Attempts will then be made to use the concatenated gene approach, combining the

18S gene, with the cox1/2 gene, to yield more accurate trees with greater powers of

discrimination. With the ultimate aim of providing a tool to improve the

phylogenetic techniques to classify the subgenus and improve the diagnosis of AK,

ideally differentiating between pathogenic and non-pathogenic strains.


2.2 Materials and Methods

2.2.1 Chemicals

Chemicals were obtained from Sigma-Aldrich (Gillingham, U.K.) unless

otherwise stated. They were all sterilised by either autoclaving or by passage

through a 0.2 µm- Acrodisc® syringe filter (Pall Life Sciences, Portsmouth, U.K.)

prior to use.

2.2.2 Organisms

All strains of Acanthamoeba used in this study are held within the culture

collection of Dr Simon Kilvington, Department of Infection, Immunity and

Inflammation, University of Leicester, Leicester, U.K (Table 2). Unique AK clinical

isolates were obtained from patients receiving treatment at Moorfields Eye Hospital,

London, U.K. Corneal scrapes were collected and kindly donated by Mr John Dart

of Moorfields Eye Hospital, London, U.K: Corneal scrapes were placed, stored and

transported on agar slopes, ready for recovery and culturing techniques to be carried

out at the University of Leicester under the guidance of Dr Simon Kilvington.

Table 2. Microorganisms used for these studies.

Species CCAP/

ATCC

Strain name T

group

Morph.

group

Source

A. astronyxis a 30137 Ray and Hayes T7 I Soil, California,

USA


A. castellanii a 1501/1a

50373

Neff/ OS4-7B T4 II Soil, California,

USA

A. comandoni a 1501/5 - - - Soil, France

A. culbertsoni a 30171 Lilly A-1 [AC-

001]

T10 III Primary monkey

kidney tissue

culture, India

A. griffinii a 1501/4 - T3 II Marine beach,

Connecticut,

USA

A. hatchetti a 30730 Bh2 T11 II Sediment,

Baltimore

Harbour,

Maryland, USA

A. healyi a - CDC1283

VO13

T12 III GAE, brain,

Barbados, BWI

A. lenticulata a 30841 PD2S T5 III Swimming pool,

France

A.

palestinensis a

30870

1547/1

AC-014 T2 III Soil, Israel

A.

palestinensis a

50708

1501/3c

OX-1/2802 T6 III Swimming pool,

France

A. polyphaga a 1501/3g - - - AK, USA

A.

palestinensis a

1547/1 - - - -


A. polyphaga a 30873

1501/3d

Nagington T4 II AK, UK

A. polyphaga a - Ros T4 II AK, UK

Acanthamoeba

sp. a

- Environmental

1

- - Cold water

storage tank, UK

Acanthamoeba

sp. c

- A Keratitis

(AK) 1 to 19

- - AK, UK

Acanthamoeba

sp. b

- AK95/1153 - - AK, UK

A. tubiashi a 30876 - T8 I Freshwater,

Maryland, USA

a Dr S. Kilvington, University of Leicester;

b Prof. D. Warhurst, London School of

Hygiene and Tropical Medicine (LSHTM); c Mr J. Dart, Moorfields Eye Hospital; -:

no information available.

2.2.3 Monoxenic culture of Acanthamoeba

Acanthamoeba trophozoites were cultured by monoxenic growth on non-

nutrient agar plates seeded with a lawn of E. coli (NNA-E. coli) (Page, 1988). The

NNA media was comprised of 1.5% plain agar (Agar No. 1, Lab M™, Bury U.K.)

and 1 tablet of ¼ strength Ringer’s solution per 500 ml of deionised water. The agar

mix was autoclaved at 121C for 15 minutes, and allowed to cool to 50C before

being poured into plates, and left to dry overnight at 37C.

Once dried, the agar surface is seeded with 2-3 drops of a dense suspension

of bacteria E. coli strain JM101 (ATCC 33876) (see 2.2.3.1). The bacteria are

spread evenly over the surface with a sterile bacteria spreader, and the plates left to


dry at room temperature. Seeded plates remain viable for two weeks, if stored at

4C.

For routine maintenance of Acanthamoeba, amoebae seeded plates were

cultured in air at 32C in sealed polythene bags. Cultures were refreshed weekly by

excising a one cm2 region of agar from the leading edge of the plaque, containing

numerous trophozoites, and placing it face down onto the centre of a fresh NNA-E.

coli plate for incubation as described above. To isolate and clone single cysts from a

mixed culture, individual cysts were transferred by micro capillary manipulation on

to a fresh NNA-E. coli, and again incubated as above.

2.2.3.1 Preparation of E. coli food source stock

To make the stock suspension, E. coli was streaked on to a Luria-Bertani

(LB) agar plate using a sterile disposable loop (Fisher Scientific UK). The agar

plates made comprises of 7.5 g plain agar (Lab M™), and 12.5 g of Difco LB

powder (BD Biosciences, Oxford, U.K.) per 500 ml of distilled water, and

autoclaved at 121C for 15 minutes, and allowed to cool to 50C before being

poured into plates, and left to dry overnight at 37C (plates are viable for 2 weeks

when stored at 4C). Once inoculated, the LB plates were incubated overnight at

37C allowing colonies to form.

Distinct single colonies were picked with a sterile disposable loop (Fisher

Scientific U.K.), and transferred to a 175 cm2 tissue culture flask (Nunc- Fisher

Scientific U.K.) with 100 ml of LB broth (comprising of 12.5 g of Difco LB powder

(BD Biosciences) in 500 ml of distilled water, and autoclaved at 121°C for 15

minutes), and propagated overnight at 37C in a shaking incubator. The suspension


of E. coli was harvested under aseptic conditions and transferred to two 50 ml

polypropylene tubes. The bacteria were pelleted by centrifugation at 2,000 x g for

30 minutes, and the supernatant discarded. Next, all culture media was removed

preventing additional multiplication of the bacteria, by resuspending and thereby

washing the pellet in ¼ strength Ringer’s solution. This wash step was carried out a

total of three times, with the final pellet being resuspended in 10 ml of ¼ strength

Ringer’s solution.

The stock suspension of bacteria as a food source can be stored at 4C for up

to two weeks.

2.2.4 Axenic culture of Acanthamoeba

Where possible, strains were adapted to axenic growth in a semi-defined

media (Hughes and Kilvington, 2001), comprising of 20 g Biosate peptone (BD), 5

g Glucose, 0.3 g Potassium dihydrogen orthophosphate (KH2PO4), 10 mg Vitamin

B12, 15 mg L-Methionine per 900 ml of double distilled water, and if necessary pH

was adjusted to 6.5-6.6 with 1 M sodium hydroxide (NaOH). The culture media was

divided in volumes of 225 ml and autoclaved at 121˚C for 15 minutes. Prior to use,

penicillin/streptomycin solution (to a final concentration of 100 U/ml and 0.1

mg/ml, respectively) was added to all the media aliquots. These were then made up

to 250 ml with either sterile distilled water, or 10% heat inactivated foetal calf

serum (Invitrogen Ltd, Paisley, U.K.) depending on the strain of Acanthamoeba

being cultured. The complete media can be stored at 4˚C, and used within one

month.

Before axenic culture could be obtained superfluous bacteria surrounding

the Acanthamoeba was removed by treating cysts with 2% (v/v) hydrochloric acid


(HCl) for 24 hours (Kilvington and White, 1994). All traces of HCl were then

removed by washing the cysts in sterile deionised water and pelleting with

centrifugation at 1,000 x g for 3 minutes, this wash step is repeated three times.

Washed cysts are inoculated into flat-sided Nunclon™ Surface tissue culture tubes

(Fisher Scientific UK) containing 3 ml of culture media, and incubated in air at

32C. Under these conditions, excystation should occur and any emergent

trophozoites can adapt and replicate in the media. Axenic strains were successfully

maintained in 25 cm2 (small) tissue culture flasks (Nunc- Fisher Scientific UK).

Cultures were maintained on a weekly basis by harvesting the amoebae

under axenic conditions and leaving a seeding stock replenished with fresh media,

this ensured cells did not become over confluent and were typically maintained at

mid log phase when metabolism is greatest. When required, cell density was

assessed microscopically using a modified Fuchs Rosenthal haemocytometer (SLS,

Nottingham, U.K.) and adjusted accordingly. When a higher density of cells was

required, a seeding stock was inoculated into an appropriately sized tissue culture

flask (Nunc- Fisher Scientific UK) under axenic conditions and incubated at as

before.

2.2.5 Cryopreservation of Acanthamoeba

As a standard procedure Acanthamoeba isolates were cryopreserved to

maintain integrity and avoid potential subculturing cross contamination.

Axenic trophozoites were grown to late log phase (approximately 1 x 106), and

then pelleted by centrifugation (500 x g, for 5 minutes). The pellet was resuspended in

(heat- inactivated) foetal bovine serum (FBS) (Gibco®, Invitrogen, Paisley, U.K.)

supplemented with 5% DMSO and 0.5 ml volumes were transferred to cryogenic vials


(Nalgene- Fisher Scientific UK). The vials were immediately placed in a

cryopreservation unit- 5100 Cryo 1C freezing container “Mr. Frosty”, (Nalgene-

Fisher Scientific UK). The commercially available unit is designed specifically for

freezing organisms in cryogenic vials, and contains a special compartment to house

propan-2-ol ensuring it surrounds the vials. The cryopreservation unit is placed on the

bottom shelf of an -80C freezer (Sanyo Gallenkamp, Loughborough, U.K.) for a

minimum of four hours. Once in the freezer the container and propan-2-ol controls the

cooling rate of the cells to 1C /minute when they are in the presence of a

cryoprotective agent commonly dimethyl sulfoxide (DMSO), therefore preventing

intracellular ice crystal formation and cellular damage.

The vials containing strains were then catalogued and plunged into liquid

nitrogen at -196C, for long-term storage.

Recovery of a cryopreserved strain was achieved by rapidly thawing the

relevant vial in a 37C water bath. Then inoculating the cells into fresh culture media

warmed to 32C; although not essential, the media was usually supplemented with

10% FBS. To reduce any toxicity caused by the DMSO the media was replaced after

six hours. After 24 hours incubation, the recovered trophozoites should have adhered

to the wall of the flat-sided Nunclon™ Surface tissue culture tube (Fisher Scientific

UK); again the culture media was replaced with a fresh supply. Confluent trophozoite

growth was usually observed after a further 48 hours incubation 32C.


2.2.5.1 Cryopreservation of bacteria

Bacteria were cryopreserved, by suspending the cells in LB broth containing

5% DMSO and rapidly freezing catalogued vials directly at -80C (Sanyo) for long-

term storage. Cells were recovered by rapid thawing at 37C, and inoculated on to

appropriate agar media or broth for incubation at 37C. Again media was replaced

after six hours to reduce any toxicity caused by the DMSO.

2.2.6 Acanthamoeba DNA isolation

Axenic Acanthamoeba cultures were harvested for high quality genomic

DNA extraction by the UNSET method (Gast et al., 1994). If attempts to axenise

strains had failed, the cells were continued to be maintained and propagated by the

slower method of monoxenic culturing (section 2.2.3), Acanthamoeba cells and

subsequently DNA was collected directly from the agar surface, with resulting

lower concentrations of recovered DNA (see section 2.2.6.1).

Cells were harvested by centrifugation at 1,000 x g for 5 minutes, and the

supernatant gently decanted. Cells were resuspended in 10 ml of ice cold

Dulbecco’s phosphate buffered saline (dPBS) (one tablet (Oxoid, Basingstoke,

U.K.) dissolved in 100 ml, autoclaved at 121˚C for 15 minutes), and transferred to a

14 ml polypropylene centrifuge tube. Once washed, the cells were pelleted by

centrifugation at 1,000 x g for 5 minutes, and the supernatant removed. 3 ml of

UNSET buffer lysis was added (8M urea, 0.15M NaCl, 2% sarkosyl, 1mM EDTA,

0.1M Tris, per 1,000 ml of nH2O, adjusted to pH 8.0 if necessary) and the solution

gently inverted for approximately 10 seconds until lysate clears. Immediately DNA

was extracted with the addition of 3 ml of phenol-chloroform (1:1), and gently


rocked from side to side ensuring the phases are completely mixed. The phases were

separated by centrifugation at 2,000 x g for 2 minutes and the lower phase removed,

using a fine tip pipette. These steps were repeated with a further 3 ml of phenol-

chloroform (1:1), and followed by 3 ml of chloroform: isoamyl alcohol (24:1). With

these steps culminating in a final spin of 2,000 x g for 15 minutes. The upper

aqueous phase then transferred to a fresh polypropylene tube, and 0.8 volumes of

ice-cold isopropanol added. The solution was gently mixed before being held at

room temperature for 30 minutes or overnight at 4˚C. All precipitated DNA was

pelleted by centrifugation at 2,000 x g for 15 minutes. The resulting pellet was

washed with 1 ml of 70% ethanol, the pellet dislodged and recentrifuged at 2,000 x g

for 5 minutes. All supernatant removed with a fine bore pipette. This ethanol wash

step was repeated twice more. After the final removal of the supernatant, all

remaining alcohol was evaporated in an air incubator at 32˚C (approximately 20

minutes) with the cap loosened. Ultimately the pellet was left to dissolve at 4˚C in

200-400 mL of TE0.1 buffer (10mM Tris HCl, 0.1mM EDTA, autoclaved at 121˚C,

and stored at 4˚C), containing RNase A (5 μl/ml) to remove any RNA.

The presence of the DNA was determined by running an aliquot on an

agarose TAE gel (see section 2.2.7): With DNA quantity obtained using GeneSnap

(version 7.08) (Syngene, Cambridge, U.K.) with the G: BOX XT gel documentation

system (Syngene).

2.2.6.1 DNA extraction from Acanthamoeba in monoxenic culture

To harvest Acanthamoeba from the surface of the NNA plates, cells must be

washed off the agar with 5 ml of ice cold dPBS, squirted repeatedly across the


surface until the amoebae cells are dislodged. The solution is then transferred to a

14 ml polypropylene tube, and centrifuged at 1,000 x g for five minutes. The

supernatant was gently decanted, and the remaining amoebae washed within the

tube with 10 ml of ice-cold dPBS and centrifuged again. This wash stage was

repeated three times, to remove any remaining E. coli, before DNA extraction could

commence (section 2.2.6).

2.2.7 Agarose gel electrophoresis

PCR product/DNA samples were separated in 1.2% 1X-Tris-acetate (TAE)-

agarose gels (TAE 10X stock solution: 48.4 g Trizma base, 11.4 ml glacial acetic

acid, 20 ml EDTA (0.5 M, pH 8.0) per 1,000 ml nH2O) at 2 v/cm (approximately

one hour) containing ethidium bromide (EtBr) 0.5 μg/ml, and visualised with a G:

BOX XT gel documentation system (Syngene).

If required, a specific band can be isolated from a reaction containing multiple

bands. Specific sized DNA fragments can be separated directly from a lower

percentage agarose gel. Certain parameters must be followed to prevent degradation

and cutting of the DNA caused by prolonged exposure to UV light. The DNA

sample is separated in a 0.8% 1X-TAE agarose gel, at 2 v/cm (approximately one

hour) containing EtBr (0.5 μg/ml). Prior to UV exposure the gel is placed on a glass

plate, and the DNA fragment of interest is excised quickly with a sterile scalpel

blade, and transferred into a 1.5 ml eppendorf tube.

The quantity of DNA was obtained using GeneSnap (version 7.08) (Syngene)

with the G: BOX XT gel documentation system (Syngene).


2.2.8 Primer design

To obtain gene sequences from such a broad range of Acanthamoeba

isolates several primer pairs were used. The JDP and 18s primers were obtained

directly from the literature, although modification was made to 18sR (SSU1) to

improve its specificity (Table 3), and the cytochrome oxidase (cox) primers were

designed specifically for this study (Table 3). Primers were identified using

GeneFisher software (Giegerich, 1996), and constructed by Eurofins MWG Operon

(Eurofins MWG Operon, London, U.K.). Following primer optimisation reactions,

basic PCR was carried out with the relevant primer specific parameters shown in

(Table 3).

2.2.9 PCR amplification of DNA

PCR amplification of DNA was performed in either a 96-well GeneAmp®

PCR System 9700 (Applied Biosystems, Warrington, U.K.) or a 24-well Perkin-

Elmer GeneAmp® PCR system 2400 (Applied Biosystems). Using either a standard

pre-made PCR reaction mixture (2X Reddymix™ PCR master mix; 1.5 mM MgCl2;

ABGene, Surrey, U.K.), or one prepared using individual components (ABGene).

All oligonucleotide primers were purchased from Eurofins MWG Operon.

To complete the pre-prepared PCR mix (2X Reddymix™; ABGene),

approximately 100 ng of genomic DNA template was added, as well as 1 μM of

both forward and reverse primers, and if necessary the MgCl2 concentration was

upwards adjusted from the initial concentration of 1.5 μM. Finally the mixture was

made up to a total volume of 50 μl PCR mix with nH2O.

The PCR mixture prepared with individual components consisted of

approximately 100 ng of genomic DNA with 1X reaction buffer, magnesium


chloride (MgCl2) (1.5 mM - 2.5 mM), 400 μM of dNTP, 1 μM of each forward and

reverse primer, 1X sucrose dye (10X: 20% sucrose, 2 mM cresol red), and 1.5 U/μl

of Red Hot DNA Polymerase (ABGene, Surrey, U.K.). The mixture was then

increased to a total volume of 50 μl with nH2O.

Thermal cycling conditions varied according to the oligonucleotides used

within the reaction, but a typical cycle comprised of an initial DNA denaturing stage

at 95˚C for 5 minutes, followed by 35 cycles of 95˚C for 30 seconds, 50˚C - 62˚C

for 30 seconds (primer annealing depending on primer Tm), followed by 72˚C for

amplification of product. The extension time was dependent on the size of the

product being produced (30 seconds per 500 bp (Sambrook et al., 1989)). Followed

by a final extension step for 10 minutes at 72˚C, however if the product was to be

cloned in a TA vector this duration of this step was increased to 30 minutes. Finally

the reaction was held at 14˚C, before a small quantity of the product was analysed

by agarose gel electrophoresis (section 2.2.7).

If PCR products of the expected size were obtained the appropriate bands

were purified by one of the techniques in section 2.2.10.

2.2.10 Purification of PCR products

As a general rule, the PEG method (2.2.10.1) was used for screening

to confirm correct band size. While for purifying DNA ready for insertion into

plasmids, the use of Microcon® centrifugal filters (device YM-50) (Millipore) were

favoured as they produced the best sequencing results and therefore the most pure

DNA of the three methods (2.2.10.1.i). If multiple DNA bands were produced by

PCR method 2.2.10.2 was used to isolate the DNA of the correct size.


2.2.10.1 PEG purification of single bands

Production of a single band allows target DNA to be precipitated directly

with a non-ionic water-soluble uncharged polymer such as polyethylene glycol

(PEG) (Cole, 1991), which can then be dissolved in nH2O. Or centrifuged in a

single-spin Microcon® centrifugal filter device (YM-50) (Millipore, Watford,

U.K.) that also results in sequencing or cloning grade DNA.

For the precipitation of DNA using PEG, 50 µl of 20% PEG (2.5 M NaCl)

(10 g PEG 8,000, 7.3 g NaCl, up to 50 ml with nH2O) was added to a 1.5 ml

eppendorf, with 45 µl of PCR mix, and mixed by repeated pipetting, before being

incubated for 15 minutes at 37C. The mixture was then centrifuged at 12,000 x g

for 15 minutes, and the supernatant carefully removed. The DNA pellet was washed

with 0.5 ml of cold (4C) 70% ethanol (EtOH), and again centrifuged at 12,000 x g

for 5 minutes. The EtOH was again carefully removed, with any remaining alcohol

allowed to evaporate during incubation at 32C. Once completely dried the purified

PCR products were dissolved in 20 µl of nH2O (by pipetting several times and

warming at 37C the process can be assisted). Recovery can be confirmed by

running 2 µl on an agarose gel (section 2.2.7).

2.2.10.1.i Micron®–PCR purification of single bands

The alternative method using Microcon®–PCR filter purification units

(Millipore), and the supplied instructions, results an average 95% recovery rate. The

system relies on an ultracel YM membrane and centrifugal force. The Microcon

sample reservoir was inserted into the vial, and the PCR solution carefully pipetted

into the reservoir (avoiding contact with the membrane), and the lid is sealed. The

assemblage was then centrifuged at 14,000 x g for 12 minutes, at 25C. Finally the


sample reservoir was removed from the original vial, and transferred inverted, to a

new 1.5 ml eppendorf, and again centrifuged, but this time at 1,000 x g for 3

minutes, collecting the now purified PCR product into the base of the new

eppendorf.

2.2.10.2 Target DNA purification from a multiple band PCR

If PCR amplification resulted in multiple bands, purification by an

alternative method is necessary. Products from the reaction mix were separated in a

0.8% agarose gel, and the target band excised with a sterile scalpel. The gel segment

containing the band was heated to 60C with 3X 7 M guanidine acetate (67 g

Guanidine hydrochloride, 20 ml Potassium acetate 3 M, pH 4.8 (29.4 g Potassium

acetate, 11.5 ml Glacial acetic acid, to 100 ml with nH2O, autoclaved at 121C for

15 minutes, and stored at 4C), total volume was increased to 90 ml and the pH

adjusted to 5.5 with NaOH (10 M) if necessary), whilst warming and stirring.

Volume is then increased to 100 ml and the solution passed through a 0.45 µM filter

membrane), until the agarose gel dissolves. The chaotropic solute guanidine acetate

increases the entropy of the system and disrupts the hydrogen bonds of DNA,

forming a hydrophobic environment. Driven by this dehydration, when salt

concentrations are high, the positively charged salt ions form a salt bridge between

the negatively charged nucleic acids and the negatively charged silica. Proteins

metabolites and other contaminants do not bind to the silica with the nucleic acids

(Boom et al., 1990, Nelson, 1992, Vogelstein and Gillespie, 1979). Accordingly 20

µl of silica in nH2O (300 mg/ml) was added to the reaction (2 g silica suspended in

20 ml nH2O, and allowed to settle for two hours. The milky supernatant removed

again and the silica resuspended in a further 20 ml of nH2O, these steps repeated

twice more, before the remaining volume of silica was estimated, and the silica re-


suspended in 2 volumes of nH2O). Ensuring it was fully in-suspension before doing

so. The PCR reaction, guanidine acetate and silica mixture was vortexed and

incubated at ambient temperature for 5 – 10 minutes with occasional further gentle

mixing. The mixture was centrifuged at 10,000 x g for 10 seconds and the

supernatant removed. The DNA bound silica was washed three times in 80%

isopropanol, spinning at 10,000 x g for 10 seconds and removing all supernatant,

between washes. Ensuring the eppendorf cap was opened, the DNA bound silica

was incubated at 37C for approximately 10 minutes ensuring the pellet was

completely air-dried. The DNA was then dissolved in 20 µl of pre-warmed nH2O,

and incubated for 5 minutes at 60C (with occasional mixing). Finally the silica was

removed by centrifuging at 10,000 x g for 1 minute, and the supernatant containing

the DNA was recovered in to a fresh eppendorf.

2.2.11 Ligation of DNA into cloning vectors

Ligation reactions were carried out in a 0.5 µl eppendorf. In a standard

reaction containing 3.5 µl of amplified cox1/2 or 18s PCR product, 5µl ligation

buffer (2X), 25 ng linearised pGEM®-T Easy vector (Promega UK, Southampton,

U.K.), 3 Weiss units of T4 DNA ligase (Promega UK), and nH2O to a final volume

of 10 µl were gently mixed by pipetting (so as not to form air bubbles). Ligation

mixtures were incubated at ambient temperature for one hour, or 48 hours at 14C.

A small aliquot of the modified plasmids were visualised and quantified by

1.5% agarose gel electrophoresis (detailed in section 2.2.7).


2.2.12 Production of ultra competent E. coli cells

Ultra competent E. coli cells were produced for transformation using a

simplified method, based on the technique developed by Inoue (Inoue, 1990), and

produces cells with an efficiency of >108 cfu/µg of plasmid DNA.

One ml of an overnight culture of E. coli DH5α (ATCC 53868) was added to

a 2 L flask containing 250 ml of super optimal broth (SOB) (20 g Bacto™ tryptone

(BD Biosciences), 5 g Bacto™ Yeast extract (BD Biosciences), 0.5 g NaCl, per 950

ml of nH2O; Dissolved by shaking, then added 2.5 ml KCl (1M), adjusted to pH 7

with NaOH (5 M), and made up to 990 ml with nH2O: Autoclaved at 121C for 15

minutes, and finished by added 5 ml MgCL2 (2 M), and MgSO4 (2 M)).

Over a period of approximately two days, the cells were incubated at 18C in a

shake incubator (200-250 rpm), until an OD600 = 0.6 (optimal but 0.4 to 1 will

work). Once the correct OD had been reached, the cells were placed on ice for 10

minutes, before being pelleted at 2,500 x g for 10 minutes at 4C.

Next the cells were carefully resuspended without air bubbles in 80 ml of ice

cold transformation buffer (TB) (3.36 g PIPES, 2.2 g CaCl2, 18.64 g KCl, per 900

ml nH2O: pH adjusted to 6.7 with KOH (5 M), whilst slowly stirring 10.88 g

manganese chloride (MnCl2) was added: Volume adjusted to 1 L with nH2O: Filter

sterilised and stored at 4C). The cells were stored on ice for 10 minutes, before

being pelleted again at 2,500 x g for 10 minutes at 4C. Cells were carefully

resuspended (air bubble free) in 20 ml ice cold TB: 7% DMSO was added and the

solution placed on ice for 10 minutes. Aliquots of 0.5-1 ml quantities were

transferred to cooled cryotubes and frozen in liquid N2. The cells can be stored for

up to a one-year at –70C.


2.2.13 Heat shock transformation of ultra competent E. coli

The fragile ultra competent E. coli were thawed slowly on ice, for

approximately 30 minutes. The solution was distributed in 200 µl aliquots, into

cooled 1.5 ml eppendorfs and placed on ice. Added to one aliquot was 100 ng of

circular pGEM®-T Easy plasmids now complete with inserts, while the control

aliquot had an addition of 2 µl of TE0.1 as a substitute for the lack of plasmid. The

tubes were incubated on ice for 30 minutes, with the occasional gentle mix. The

incubated solutions were exposed to a heat pulse without agitation at 42C for 45

seconds, and transferred to an ice bath for 2 minutes. Next is the addition of 0.8 ml

of SOC (SOB, with 20 mM filter sterilised 1 M glucose (super optimal broth with

catabolite repression)), and the cells are left to recover and multiply in a shaking

incubator (200-250 rpm) at 37C.

Following incubation, identification of successful transformants is carried

out on specific selective LB agar plates. A 200 µl aliquot of the transformation mix

and control were plated on to LB agar indicator plates containing ampicillin (100

μg/ml), which have been surface supplemented with the blue/white screening

indicator solutions, X-gal- 5-bromo-4-chloro-3-indolyl-β-D-galactoside in

dimethylformamide (80µg/ml), and Isopropylthio-β-D-galactoside in nH2O (IPTG)

(0.5mM): These solutions must be added respectively, with each being left for a

period of time to allow the liquid to evaporate. Once the surface is dry, the plates

can be inoculated with transformed bacteria.

When antibiotic pressure is required, the agar is heated until molten but

allowed to cool to 50C before the addition of the required quantity of antibiotic.

The bottle is then gently shaken to ensure even distribution throughout the solution.


In addition, the remaining solution from the test transformation mix was

centrifuged, and the supernatant removed. The resultant pellet of cells was also

plated on to an LB agar ampicillin indicator plate. All seeded plates were incubated

overnight at 37C.

Multiple second round selections on selective indicator LB agar were made

of successful transformants to ensure single colonies are isolated. From this second

plate distinct colonies were selected for cultivation in LB broth (section 2.2.3.1)

with continued selective pressure of ampicillin at 100µg/ml, in a shake culture of

37C, revolving at 180 rpm.

2.2.14 Restriction enzyme digestion

In a 0.5 ml eppendorf, the following components were mixed to produce a

2X restriction enzyme digestion, 2 µl of nH2O, 1 µl of restriction enzyme EcoR 1

(Promega UK) and 2 µl of 10X enzyme buffer H (1X: 90mM Tris-HCl, 10mM

MgCl2, 50mM NaCl, pH 7.5) (Promega UK) (8-12 U/µl), (the enzyme must be kept

on ice until returned to -20C). Added to the reaction mixture was 15 µl of ligated

target DNA (pGEM®-T Easy plasmids with insert). The mixture was incubated for

4-12 hours, at 37C, and a small amount of the reaction was resolved by agarose gel

electrophoresis (as in section 2.2.7).


Table 3. Primers (5' to 3') used for gene sequence typing of Acanthamoeba spp. SSU 18S rDNA and cox1/2, with corresponding PCR

parameters.

Primers Sequence MgCl2

conc. MM

Annealing

temp. C

Extension

time sec.

Product

size bp

Paper cited

Acanthamoeba 18S:

JDP1F

JDP2R

GGCCCAGATCGTTTACCGTGAA

TCTCACAAGCTGCTAGGGGAGTCA

1.5 60 45 450 (Schroeder et

al., 2001)

AC-892c GTCAGAGGTGAAATTCTTGG Sequencing primer

18sF

18sR

CTGGTTGATCCTGCCAG

Originally called SSU1

GATCCTTCTGCAGGTTCACCTAC

Modified from SSU1

1.5 48 90 1.5-2 kb (Weekers et

al., 1994)

Modified for

this study

18s588 CGCGCAAATTACCCAATC Sequencing primer

Acanthamoeba cox1/2:

CoxA125 ATGATTGGHGCTCCNGAYATGG 2.5 60 45 748 Designed for


CoxA873 TGRCCTCCCCATAATGTAGC this study *

Cox1/2F

Cox1/2R

GAATTAGCTGCTCCGGGTTC

TCAGGATAATCGGGGATCCTTC

2 – 2.5 46 or 52 90 1.2 kb (Kilvington et

al., 2004)

CoxA-486F

CoxA-

1057R

GCHGGTGCTATTACTATGCTTT

CCWGCAAARAARGCAAAAACDGC

1.5 56 (60

touchdown)

30 566 Designed for

this study *

SeqCoxA GGTGCTATTACTATGC Sequencing primer

*Wobble nucleotides for degenerate primers were those according to Operon: H, A/C/T; W, A/T; R, A/G; N, A/T/G/C; Y, C/T.


2.2.15 Plasmid purification

Several distinct white colonies were chosen at random from the LB agar

(ampicillin 100 μg/ml) indicator plates, seeded with pGEM®-T Easy (plus insert)

plasmids and transformed E. coli. A small section of each colony was inoculated

into 5 ml universal tubes containing LB broth (ampicillin 100 μg/ml), and

propagated overnight in a shaking incubator at 37C, revolving at 180 rpm.

Plasmids were then purified with the relevant method depending on their

downstream application.

For quick screening to confirm insert presence and size, the silica method

(section 2.2.15.i) was selected. If plasmids were required for sequencing, either the

comprehensive method (section 2.2.15.ii), or the plasmid purification kit (Qiagen

Ltd, Crawley, U.K.) (2.2.15.iii) was carried out. As a general rule the

comprehensive method was trialled first, but if failed at the sequencing stage as a

result of impurities, plasmids were then purified using the commercial kit on the

second attempt.

Once screening of the RE digests confirmed which colony contained a

plasmid with the correct size insert, it was selected and propagated in an LB broth

shake culture, before being purified for sequencing (as in section 2.2.15.iii).

2.2.15.1 Silica method of plasmid purification for screening

From an overnight culture of transformed E. coli DH5α, 1.4 ml was

transferred into a 1.5 ml eppendorf, and pelleted at 6,000 x g for 2 minutes.

Bacterial cell membranes were broken down and chromosomal DNA precipitated

by the addition of lysis buffer, and GTE. The supernatant was removed, and the


remaining pellet resuspended in 120 l GTE (50 mM Glucose, 25 mM Tris pH 8, 10

mM EDTA pH 8). Followed by the addition of 240 µl of lysis buffer (200 mM

NaOH, 1% SDS, (Stored for up to one week at 4C)). The tube was inverted six

times, and incubated on ice for 5 minutes. Then 360 µl KAc (adjusted to pH 5.5

with acetic acid) was added, and the tube inverted six times, followed by

centrifugation at 10,000 x g for 10 minutes. The supernatant was transferred to a

new eppendorf, leaving behind any bacterial cell membranes.

To the new tube 20 µl of fully resuspended silica suspension was added

(2.2.10.2), and the solution vortexed. The now DNA bound silica was pelleted by

centrifugation at 10,000 x g for 10 seconds and all the supernatant removed. The

silica was resuspended in 500 µl of wash solution (50 mM NaCl, 10 mM Tris pH

7.5, 2.5 mM EDTA, 50% ethanol) by gently pipetting. Once washed, the silica was

again pelleted at 10,000 x g for 10 seconds, and the supernatant removed. A second

spin was carried out, and any residual wash buffer removed. The silica was left to

fully dry in their tubes with the caps open, in an incubator at 37C for

approximately 10 minutes. Finally the DNA was eluted into nH2O, by resuspending

the silica in 50 µl nH2O and incubating at 60C for 5 minutes (with occasional

mixing). To remove the silica from the DNA in suspension, the solution was

centrifuged at 12,000 x g for 1 minute, and the supernatant transferred to a new

eppendorf.

Purified plasmids were visualised and quantified in 1.5% agarose gel

electrophoresis (detailed in 2.2.7).


2.2.15.2 Comprehensive plasmid purification for sequencing

From an overnight culture of transformed E. coli DH5α, 1.4 ml was

transferred into a 1.5 ml eppendorf, and pelleted at 12,000 x g for 1 minute, and the

supernatant removed. Bacterial cell membranes were broken down and

chromosomal DNA, precipitated by the addition of lysis buffer, and GTE. The

supernatant was removed, and the remaining pellet resuspended in 200 µl GTE (50

mM Glucose, 25 mM Tris pH 8, 10 mM EDTA pH 8). Followed by the addition of

300 µl of lysis buffer (200 mM NaOH, 1% SDS), (Stored for up to one week at

4C)). The solutions were mixed by tube inversion, and incubated on ice for exactly

5 minutes. A further addition of 300 µl KAc (adjusted to pH 5.5 with acetic acid)

was carried out, and again mixed by inversion, and followed with a 5 minute

incubation on ice. The solutions were then centrifuged at 12,000 x g for 10 minutes,

and the supernatant transferred to a new eppendorf (between 700-750 µl). Rnase A

to a final concentration of 20 µg/ml was added, and the solution left to incubate at

37C for 20 minutes.

To remove unwanted proteins and oligosaccharides two chloroform

extractions were carried out. Here 400 µl of chloroform was added, and the

solutions mixed thoroughly. Centrifuging at 12,000 x g for 1 minute separated the

phases, and the upper aqueous layer removed to a clean eppendorf. Two chloroform

extractions were completed, and then DNA precipitated by adding equal volumes of

isopropanol, and incubating on ice for 10 minutes. All DNA was pelleted by

centrifugation at 12,000 x g for 15 minutes at ambient temperature. The surface of

the pellet was then washed with 500 µl of 70% ethanol. All traces of the alcohol

was removed, following a spin at 12,000 x g for 2 minutes, and air drying at 37C

for approximately 30 minutes. The DNA pellet was dissolved in 32 µl of nH2O.


Once dissolved, 8 µl of NaCl (4 M) and 40 µl of 13% PEG 8,000, was added

and thoroughly mixed, then incubated on ice for at least 20 minutes. The DNA was

again pelleted by spinning at 12,000 x g for 15 minutes at 4C. The supernatant

removed, and the now translucent pellet washed in 70% ethanol, this step was

carried out twice with 12,000 x g centrifugation between each wash. Following the

washing steps all traces of ethanol was removed, after centrifugation at 12,000 x g

for 5 minutes, and the pellet left to air dry at 37C. Ultimately the pellet was

dissolved in 25 µl nH2O.

Purified plasmids were visualised and quantified by 1.5% agarose gel


2.2.15.3 QIAGEN® plasmid maxi kit

The comprehensive method for purifying plasmids for sequencing was used

initially but replaced by QIAGEN® plasmid maxi kits (QIAGEN Ltd), also based

on alkaline-lysis, and followed by binding of the free DNA to anion-exchange resin,

therefore recovering the plasmids, before dissolving them in nH2O.

Purified plasmids were visualised and quantified by 1.5% agarose gel


2.2.16 Sequencing

The inserts were commercially sequenced using the standard technique of

automated sequencing at either the Protein Nucleic Acid Chemistry Laboratory

(PNACL; University of Leicester, U.K.) using fluorescently labelled M13 or T7

primers in a 3730 DNA analyser automated sequencer (Applied Biosystems,


Warrington, U.K.), or at Eurofins MWG Operon using their value read sequencing

service.

2.2.17 Sequence analysis

Sequence data was viewed on a chromatogram using Chromas software

version 2.23 (http://www.technelysium.com.au/chromas.html); primer sites were

excluded from analysis to reduce any forced bias. Sequences were then identified

via a basic local alignment search tool (BLAST) (Altschul et al., 1997), using

nucleotide blast (blastn) to locate other highly similar sequences for comparison.

Multiple sequences of the same target region from different strains and species were

aligned together using ClustalW (Larkin M.A. et al., 2007), through the BioEdit

interface (version 7.0.4.1.) (Hall, 1999). With BioEdit, sequences can be aligned

(ClustalW) and extraneous data including primer-binding sites can be easily

removed, which removes any bias caused by their inclusion in the alignment. The

software ClustalW is a general purpose multiple sequence alignment program for

DNA, which aligns the sequences allowing single nucleotide differences to be

identified. Finally phylogenetic and molecular evolutionary analyses were

conducted using MEGA version 5.05 (Kumar et al., 2004), including the

concatenation of the two genes: The multiple aligned sequences were subject to

phylogenetic reconstruction to produce a gene tree using neighbour-joining (NJ)

distance analysis or maximum parsimony (MP) with the kimura two-parameter

correction, and bootstrap support for all phylogenetic trees determined from 1000

replications.


2.3 Results

2.3.1 18S genotyping of strains

Sequences of 18S rDNA and mt cox1/2 were obtained from 28 unique drug

resistant strains, isolated by corneal scrapings from 17 patients. These strains were

added to a range of 12 known isolates, including type species, from all three

morphological groups and 11 T-groups. Sequence variation across ~204 bp and

~564 bp respectively, from the group of 40 isolates was analysed.

Phylogenetic relationships among isolates were examined by maximum

parsimony (MP) and neighbour joining (NJ) analysis, and comparisons were made

of any group’s formed, bootstrap values, and tree topography.

Bootstrap values have been included to help determine groups, as it is

generally considered that values of greater than 70 are evidence to support the

distinction of the clade. Phylogenetic trees produced by both NJ and MP algorithms,

do vary from each other (NJ: <48 = 35%; 49-94 = 50%; >95 = 15%. MP: <48 =

86%; 49-94 = 11%; >95 = 3%), with NJ trees resulting in a greater number of

higher bootstrap values. However the clusters of taxa remain largely the same,

although their position within the tree vary depending on the analysis used.

The 18S NJ tree (Figure 1) shows multiple species are clumped together in

genotype groups T3, T4 and T11, with the maximum pairwise distance value

between the three clades at 0.055 (T11 verses T3), and a minimum value of 0.033

(T3 verses T4 (Table 4)). The 18S tree highlights the close relationship between T3,

T4 and T11 clades. With the T11 group located between two subsections of T3,

where isolates AK 1a, AK 4, AK 6 and AK 14 are above the T11 cluster, and AK 5,

AK 5a, AK 11 and A. griffinii 1501/4 are below (Figure 1). Also located a separate


branch of its own is the T4 isolate A. castellanii 1501/1a, which when analysed here

clusters above T11 and below the upper T3 subsection.

Table 4. Estimates of evolutionary divergence over 18S sequence pairs between T-

group genotypes. The number of base substitutions per site from between sequences





(Tamura et al., 2011) in press.

T11 T3 T4 T6 T2 T10 T9 T8 T5

T3 0.055 - - - - - - - -

T4 0.034 0.033 - - - - - - -

T6 0.104 0.081 0.084 - - - - - -

T2 0.123 0.109 0.111 0.034 - - - - -

T10 0.101 0.118 0.093 0.14 0.161 - - - -

T9 0.221 0.236 0.219 0.276 0.29 0.229 - - -

T8 0.258 0.211 0.229 0.263 0.274 0.25 0.118 - -

T5 0.091 0.091 0.072 0.079 0.07 0.1 0.229 0.249 -

T7 0.256 0.199 0.226 0.238 0.271 0.272 0.109 0.07 0.248

When comparing average pairwise distances of all the T-groups, maximum

values occur between T2 verses T8 (0.274), and minimum between T3 and T4

(0.033) (Table 4). Differences between the closely related sequence types; T3, T4


and T11 in this study, are less than the previously recorded value of at least 5%

(which is an arbitrary value) (Stothard et al., 1998). Here, dissimilarity values of

3.3% (T4 verses T3), 5.5% (T3 verses T11) and 3.4% (T4 verses T11) were

obtained. Sequences types T2 and T6 also showed high sequence similarity, with a

dissimilarity score of 3.4%. Differences between genotypes were always greater

than within sequence types.

Of the 28 clinical AK isolates, 11 strains collected from eight patients were

analysed and assigned to the T4 clade (Figure 1). While seven AK isolates from six

patients were assigned to the clade T3 with A. griffinii 1501/4. The final 11 isolates

from six patients were included on three branches within T11 with A. hatchetti Bh2.

Distinct sequence types placed at the very bottom of the tree are all

morphological group I species, which have previously been assigned to T7 (A.

astronyxis 30137), T8 (A. tubiashi 30876), and T9 (A. comandoni 1501/5). Placed

between the lower morphological group I species, are the closely linked T2 and T6

species (A. palestinensis 1501/3c and A. palestinensis 1547/1), along with T9 A.

lenticulata PD2S on a separate branch.


AK 17

A. polyphaga Ros

AK 15

AK 13

AK 12a

AK 12

AK 7a

AK 6a

AK 7

AK 9

AK 9a

AK 18

T4

AK 1a

AK 4

AK 6

AK 14

T3

T4 A castellani 1501/1a

AK 10a

AK 19

AK 10

AK 1

AK 8a

AK 3

AK 16

AK 16a

AK 3i

AK 3a

AK 3b

A. hatchetti Bh2

T11

AK 5

AK 5a

AK 11

A. griffinii 1501/4

T3

T5 A. lenticulata PD2S

T6 A. palestinensis 1547/1

T2 A. palestinensis 1501/3c

T10 A. culbertsoni 30171

T9 A. comandoni 1501/5

T8 A. tubiashi 30876

T7 A. astronyxis 3013790

100

88

76

98

95

70

76

45

64

60

68

42

32

43

24

22

50

37

61

0.02


Figure 1. Neighbour joining distance tree based on partial 18S rDNA sequences of

Acanthamoeba spp. The tree is unrooted and obtained by Kimura two-parameters


reference bp from 1,175 to 1,379. The scale bar represents the corrected number of

nucleotide substitutions per base using Kimura method. Designated T-groups are

shown.


T6 A. palestinensis 1547/1

T2 A. palestinensis 1501/3c

T3 AK 14

T3 AK 6

T3 AK 4

T3 AK 1a

AK 9a

AK 9

AK 18

T4

AK 15

AK 17

AK 13

AK 12

AK 12a

AK 7

AK 6a

A. polyphaga Ros

T4

T4 AK 7a

A. griffinii 1501/4

AK 11

AK 5

AK 5a

T3

T4 A castellani 1501/1a

AK 10

AK 19

AK 10a

AK 1

AK 8a

AK 3b

AK 16a

AK 3i

A. hatchetti Bh2

AK 3

AK 16

AK 3a

T11

T5 A. lenticulata PD2S


T9 A. comandoni 1501/5

T8 A. tubiashi 30876

T7 A. astronyxis 30137

34

96

15

21

72

15

28

3

2

1

2

5

23

49

1

27

17

2

0

0

0

0

1

3

69

1

2

13

12

1

0

1

2

14

36

99

56



of Acanthamoeba spp. The tree is unrooted and obtained by Kimura two-parameters


reference bp from 1,175 to 1,379. Designated T-groups are shown.

2.3.2 Typing strains with cox1/2 gene

Sequences of mt cox1/2 were obtained from 28 unique drug resistant strains,

isolated by corneal scrapings, from 17 patients. These strains were added to a range

of 12 known isolates, including type species, from all three morphological groups

and 11 T-groups. Sequence variation across ~564 bp, from the group of 40 isolates

was analysed.

Phylogenetic relationships among isolates were examined by maximum

parsimony (MP) (Figure 3) and NJ analyses (Figure 4).

When examined by NJ analysis as with cox1/2, A. tubiashi 30876 (T8) was

found to be distinct from the main group of Acanthamoeba sp., but interestingly

even more distinct was the species A. hatchetti Bh2 (Figure 4). By removing these

two distinct species from the comparison, the branches of the tree become more

elongated (data not shown). Unfortunately, despite continued efforts, mt cox1/2

sequences could not be obtained for Hartmannella or Balamuthia spp. in order that

they be included as an outlying species or true evolutionary ancestor, and allow A.

tubiashi and A. hatchetti to remain within the main section of the tree.

Figure 3, displays phylogenetic relationships when examined by maximum

parsimony and includes bootstrap values to help determine clades, as it is generally

considered that values of greater than 70 are evidence to support the distinction of

the clade. Phylogenetic trees produced by both NJ and MP algorithms, do not vary


extensively from each other (NJ: <48 = 32%; 49-94 = 42%; >95 = 19%. MP: <48 =

50%; 49-94 = 29%; >95 = 18%), and the clusters of taxa remain the same, although

their position within the tree may vary depending on the analysis used. The outlying

distinct taxa change, from A. tubiashi 30876 (T8) and A. hatchetti Bh2 (T11) with

NJ, to group I species A. astronyxis 1534/1 with MP analysis.

Using cox1/2 tree analysis with bootstrap values, eight groups can be

identified (Figure 3 and 4) (Group A-H). Across these sequence types, the

pathogenic strains including the Moorfield AK strains, are resolved into smaller

clades, than the general clumping together formed by 18S T-group analysis. T11

isolates are no longer clustered together, especially relevant to the repeat isolates

from Patient 3 (AK 3 AK 3i, AK 3a, AK 3b, AK 16, and AK 16a), which were all

located in a tight clade by 18S (Figure 1 and 2), but are now distributed throughout

cox1/2 groups A, C, E, G and H. Within this study A. castellanii 1501/1a, AK 10a,

A. palestinensis 1547/1, A. griffini 1501/4, A. tubiashi 30876, A. hatchetti Bh2, A.

astronyxis 1534/1, A. comandoni 1501/5 and A. palestinensis 1501/3c cannot be

assigned to a clade with a relevant bootstrap value of over 70.

Strains with pathogenic potential are found throughout the tree, and AK and

encephalitis causing strains are not distinct from those which are thought be non-

pathogenic. Interestingly, cox1/2 groups were not limited to repeat isolate sequences

but often contain isolates collected from different patients/locations. The range of

pairwise distance values between the cox1/2 sequence groups varies from a

minimum of 0.036 (group B verses C), to a maximum of 0.286 (group H verses F)

(Table 5).


Distinct species at the base of the cox1/2 trees differs depending on analysis.

By NJ A. hatchetti Bh2 and A. tubiashi 30876 are the distinct strains, but by MP

analysis the outlying species is A. astronyxis 1534/1. Sequence similarity between

the cox1/2 groups is always higher than 5%, and the differences between sequence

types is always greater than within sequence types.

Table 5. Estimates of evolutionary divergence over cox1/2 sequence pairs between

groups (A-G). The number of base substitutions per site from between sequences






G C E H B A D

C 0.229 - - - - - -

E 0.23 0.101 - - - - -

H 0.23 0.228 0.224 - - - -

B 0.189 0.036 0.09 0.193 - - -


A 0.219 0.039 0.092 0.233 0.023 - -

D 0.242 0.095 0.105 0.247 0.07 0.085 -

F 0.265 0.201 0.222 0.286 0.172 0.207 0.237


AK 12 (T4)

AK 12a (T4)

AK 3a (T11)

AK 19 (T11)

AK 3i (T11)

A. polyphaga Ros (T4)

Gp A

AK 9 (T4)

AK 9a (T4)

AK 15 (T4)

AK 17 (T4)

Gp B

AK 10 (T11)

AK 1a (T3)

AK 6 (T3)

AK 6a (T4)

AK 13 (T4)

Gp C

A. palestinensis 1547/1 (T6)

AK 14 (T3)

AK 4 (T3)Gp D

AK 7 (T4)

AK 7a (T4)

AK 16 (T11)

AK 18 (T4)

Gp E

A. castellani 1501/1A (T4)

AK 10a (T11)

A. culbertsoni 30171 (T10)

A. lenticulata PD2S (T5)Gp F

A. comandoni 1501/5 (T9)

A. palestinensis 1501/3c (T2)

A. hatchetti Bh2 (T11)

A. tubiashi 30876 (T8)

AK 8a (T4)

AK 1 (T11)

AK 16a (T11)

Gp H

A. griffinii 1501/4 (T3)

AK 5 (T3)

AK 11 (T3)

AK 3b (T11)

AK 5a (T3)

AK 3 (T11)

Gp G

Gp G A. astronyxis 1534/1 (T7)

84

93

12

100

81

63

98

99

26

66

60

100

44

10

4

11

87

41

78

32

100

14

61

15

2

16

21

89

96

35

17

24

66

21

14

66

100


Figure 3. Maximum parsimony tree based on partial mt cox1/2 sequences of

Acanthamoeba spp, with cox1/2 groups (A-H) and T-groups (T2-T11) included. The



Bootstrap values have been included, based on 1,000 bootstrap values, and are

placed at the nodes they apply to.


AK 3i (T11)

AK 3a (T11)

AK 19 (T11)


AK 12 (T4)

AK 12a (T4)

Gp A

AK 9 (T4)

AK 9a (T4)

AK 15 (T4)

AK 17 (T4)

Gp B

AK 10 (T11)

AK 6 (T3)

AK 6a (T4)

AK 1a (T3)

AK 13 (T4)

Gp C


AK 14 (T3)

AK 4 (T3)Gp D

AK 7 (T4)

AK 7a (T4)

AK 16 (T11)

AK 18 (T4)

Gp E

AK 10a (T11)



A. culbertsoni 30171 (T10)

A. lenticulata PD2S (T5)Gp F



A. astronyxis 1534/1 (T7)

AK 5 (T3)

AK 11 (T3)

AK 5a (T3)

AK 3 (T11)

AK 3b (T11)

Gp G

AK 1 (T11)

AK 8a (T4)

AK 16a (T11)

Gp H



99

99

62

79

74

96

74

94

85

80

93

57

99

12

14

11

21

22

98

39

90

31

15

85

96

5633

2566

97

63

0.05


Figure 4. Neighbour-joining distance tree based on partial mt cox1/2 sequences of

Acanthamoeba spp with cox1/2 groups (A-H) and T-groups (T2-T11) included. The



The scale bar represents the corrected number of nucleotide substitutions per base

using Kimura method.


2.3.3 Introns, multiple alleles and mixed infections

Using repeat patient isolates evidence presented here suggests two patients

contained species with multiple alleles.

Repeat samples include AK 1 and AK 1a from Patient one, AK 6 and AK 6a

from Patient six, and AK 3, AK 3i, AK 3a, AK 3b, AK 16 and AK 16a from Patient

3. 18S NJ and MP tree analysis shows Patient one’s isolate AK 1 in T11 clustering

independently from the second isolate, AK 1a in T3. The T3 clade also contains an

isolate from Patient six (AK 6a). While the other Patient six isolate, AK 6 clusters

within T4 (Figure 1 and 2).

However by sequencing a second gene, cox1/2, Patient six’s isolates AK 6 and

AK 6a, are found to have sequence homology with a pairwise distance score of 0

(cox1/2 group c by NJ and MP trees) (Figure 3 and 4), there by indicating multiple

alleles of 18S but not cox1/2.

This was not so for Patient one’s isolates (AK 1 and AK 1a), which showed

sequence variation occurring in both cox1/2 and 18S genes. The high dissimilarity

sequence variation of cox1/2 10.2% (pairwise value of 0.102) and 18S (pairwise

value 0.052) genes, suggests the presence of a mixed infection (Figure 1, 2, 3 and

4).

Isolates AK 3, AK 3i, AK 3a, AK 3b, AK 16 and AK 16a are all repeat strains

collected from Patient three. When aligned by 18S they are 100% identical, and

cluster tightly together on the same branch within T11, with A. hatchetti Bh2

(Figure 1 and 2). However by cox1/2, analysis of the data suggests that there are

four copies of the gene within the strain. The isolates are spread out across the tree

in five groups, A, C, E, G and H (Figure 3 and 4), with sequence dissimilarity

values ranging from 4.6 to 10.2% (Table 6).


Within the region of the sequenced region of the cox1/2 gene, analysis

shows no indication of introns, while evidence for the presence of multiple alleles

and mixed infections were found.

Table 6. Pairwise distance values between cox1/2 sequences of repeat isolates from

Patient three. The number of base substitutions per site from between sequences are

shown. Analyses were conducted using the Maximum Composite Likelihood model

(Tamura et al., 2004). The analysis involved 40 nucleotide sequences. All positions

containing gaps and missing data were eliminated. There were a total of 336

positions in the final dataset. Evolutionary analyses were conducted in MEGA5


AK 3 AK 16 AK 16a AK 3i AK 3a

AK 16 0.098 - - - -

AK 16a 0.098 0.100 - - -

AK 3i 0.094 0.046 0.102 - -

AK 3a 0.094 0.046 0.102 0.000 -

AK 3b 0.000 0.098 0.098 0.098 0.094

2.3.4 Concatenated sequence data

Trees were also analysed from 18S and cox1/2 data, which had been

concatenated together to determine which resulting tree from all three data sets

provided the greatest resolution.

MP and NJ trees both contained 37 nodes supported by bootstrap values.

The tree from NJ analysis (Figure 6) contains more bootstrap values of greater than

49 compared with MP trees (Figure 5) (NJ- <48 = 24%, 49-94 = 46%, >95 = 32%;


MP- <48 = 32%, 49-94 = 46%, >95 = 16%). Both contain similar clade formation,

with differing distinct strains at the base of the trees, following trees derived from

cox1/2 data alone. The distinct species by NJ are A. tubiashi 30876, and A.

hatchetti, which by MP form their own group between cox1/2 groups E and H.

Leaving A. astronyxis 1534/1 to be the distinct species in concatenated MP trees.

Another distinction between MP and NJ trees with concatenated data is the split of

group B, with AK 15 and AK 17 in the top half, and AK 9 and AK 9a in the second

half (Figure 5).

Clade formation of concatenated trees most resembles those of cox1/2, rather

than 18S. However shorter branch lengths are found on the concatenated tree than

those of cox1/2. The concatenated tree provides more resolution between species,

rather than the clumping found with 18S trees, supported by a greater number of

nodes with significant bootstrap values of greater than >95.


AK 3i (T11)

AK 19 (T11)

AK 3a (T11)

AK 12 (T4)

AK 12a (T4)


Gp A

AK 15 (T4)

AK 17 (T4)Gp B

AK 9 (T4)

AK 9a (T4)Gp B

AK 10 (T11)

AK 1a (T3)

AK 6 (T3)

AK 6a (T4)

AK 13 (T4)

Gp C

AK 4 (T3)

AK 14 (T3)Gp D


AK 7 (T4)

AK 7a (T4)

AK 16 (T11)

AK 18 (T4)

Gp E




AK 16a (T11)

AK 1 (T11)

AK 8a (T11)

Gp H

AK 10a (T11)


A.culbertsoni 30171 (T10)

A lenticulata PD2S (T5)Gp F



AK 5 (T3)

AK 11 (T3)

AK 5a (T3)

AK 3 (T11)

AK 3b (T11)

Gp G

Gp G A. astronyxis 1534/1 (T7)

67

82

99

99

99

30

99

93

83

68

72

86

89

56

99

60

54

98

17

34

50

41

65

20

68

13

78

21

13

31

60

11

5

22

27

70

89


Figure 5. Maximum parsimony distance tree based on concatenated partial mt

cox1/2 with 18S sequences of Acanthamoeba spp with cox1/2 groups (A-H) and T-

groups (T2-T11) included. The tree is unrooted and obtained by Kimura two-


based on reference bp 8,002 to 8,566.


AK 12 (T4)

AK 12a (T4)


AK 19 (T11)

AK 3i (T11)

AK 3a (T11)

Gp A

AK 9 (T4)

AK 9a (T4)

AK 15 (T4)

AK 17 (T4)

Gp B

AK 10 (T11)

AK 1a (T3)

AK 6 (T3)

AK 6a (T4)

AK 13 (T4)

Gp C


AK 4 (T3)

AK 14 (T3)Gp D

AK 16 (T11)

AK 18 (T4)

AK 7 (T4)

AK 7a (T4)

Gp E

AK 10a (T11)



A.culbertsoni 30171 (T10)

A lenticulata PD2S (T5)Gp F



AK 3 (T11)

AK 3b (T11)

AK 5 (T3)

A. astronyxis 1534/1 (T7)

AK 5a (T3)

AK 11 (T3)

Gp G

AK 16a (T11)

AK 1 (T11)

AK 8a (T11)

Gp H



100

77

100

99

45

37

79

87

100

74

62

97

64

99

9932

32

49

30

36

99

54

99

96

97

85

94

35

52

4479

3396

77

75

72

88

0.1


Figure 6. Neighbour-joining distance tree based on concatenated partial mt cox1/2

with 18S sequences of Acanthamoeba spp with cox1/2 groups (A-H) and T-groups

(T2-T11) included. The tree is unrooted and obtained by Kimura two-parameters


reference bp 8,002 to 8,566. The scale bar represents the corrected number of

nucleotide substitutions per base using Kimura method.


2.4 Discussion

2.4.1 Genotyping Acanthamoeba sp. with 18S sequences

There are many examples of the use of 18S based molecular studies for

epidemiological typing, to uncover relationships between patients with

Acanthamoeba infections and their environments (Maghsood et al., 2005, Schroeder

et al., 2001, Walochnik et al., 2000). As all forms of life share features of structure

and sequence within rRNA, analysis of it has been used to map phylogenetic history

across organisms at all levels, including the discovery of a third kingdom the

archaea, in addition to prokaryotes and eukaryotes (Woese et al., 1990).

Within Acanthamoeba, rRNA genes show a relatively slow rate of change,

and are a suitable candidate to base a genotyping system on (Gast et al., 1996).

From the similarities of the genotype clusters found by analysis of both nuclear

(Rns) and mitochondrial (rns) small subunit rRNA genes, it can be assumed that

phylogenetic clades based on 18S sequences represent evolutionary history (Ledee

et al., 2003). Thereby inferring all clades linked with AK (T2, T3, T4, T5, T6 and

T11), and GAE (T1, T4, T10 and T12), each share evolutionary adaptations to have

the potential to cause infections. In general T4 genotypes are responsible for most

AK infections, such a marked phylogenetic localisation suggests either, innate

pathogenic potential, a peculiarity of the gene sequence, or increased prevalence in

the environment (Ledee et al., 2003). Yet evidence has been presented to show the

close relationship between an eye and a lung isolate when examined by 18S

sequence, leading the authors to conclude that any pathogenic strain may be capable

of infecting more than one type of tissue (Gast et al., 1996).

By sequencing the 18S gene of Acanthamoeba it has been shown that most

isolates, both environmental and clinical including those from all forms of


acanthamebiasis can be typed into the T4 genotype (Gast et al., 1996, Schroeder et

al., 2001, Stothard et al., 1999, Stothard et al., 1998). In a single study examining

249 isolates, 179 of them belonged to T4. These isolates were obtained from both

environmental and clinical sources from, AK patients and non-AK including CSF,

brain, skin and lungs (Booton et al., 2005).

Isolates of the T4 genotype appear to be the predominant cause of AK and

GAE, and the only genotypes yet to be linked to human disease are T7, T8, T9, T13,

T14 and T15 (Khan, 2009). To find pathogenic strains classified to groups other

than T4 is less common (Booton et al., 2005, Gast et al., 1996, Stothard et al., 1999,

Stothard et al., 1998). However, exception has been shown in this study, with many

of the isolates belonging to genotypes T3, T4 and T11, but as this study group

contains largely AK causing strains it presents a strong bias towards pathogenic

genotypes. From the clinical isolates collected from 17 patients, eight of the patients

had isolates belonging to T4 (11 strains); six patients had isolates belonged to T3

(seven strains), and five patients were infected with T11 isolates (11 strains). Eleven

of the possible 15 T-groups were included in this study, and of these, three belonged

to non-pathogenic genotypes: As expected they clustered close to each other at the

bottom of the tree, distinct from the main taxa, and formed clades T7, T8 and T9.

Despite tree topology mainly following genotypes, with the exception of the split

T3 group average sequence divergence between the genotypic clades T3, T4 and

T11 based on pairwise sequence comparison are lower than the current arbitrary cut

off value of 5% (T3-T4: 3.3%; T4-T11: 3.4%), (Gast et al., 1996). The close

association between isolates from the three genotypes (T3/4/11) shown by pairwise

sequence comparisons may explain why the in both MP and NJ trees T3 isolates are

split into smaller clusters.


Although the 18S typing system has been used here to effectively classify a

unique group of AK causing isolates, and consequently confirm the robustness of

the current system. As expected, the system was not effective at resolving the

species and provided little definition between them. Instead multiple isolates were

clumped into a few large groups. A proposal has been put forward to rename

Acanthamoeba strains by T-group rather than species name, saving confusion

caused as a result of the historical classification system (Stothard et al., 1998).

However given the clustering of many species into few clades by the T-group

system, such renaming would most likely cause even more confusion to an already

complicated taxonomic system. Especially given the close association between

certain clades, where T2 and T6 have recently been referred to within the literature

as T2-T6 clade (Corsaro and Venditti, 2010, Huang and Hsu, 2010, Niyyati et al.,

2009).

Even though partial 18S sequences were obtained from all study isolates,

once subjected to phylogenetic analysis, distance trees with comparable topology to

previous studies were produced. Amplification of a shorter sequence both increases

the throughput of isolates, and reduces the potential for taq errors, without the loss

of vital data for analysis.

With the recent increase in AK incidence, and the latest outbreak in the

USA, more and more cases are being diagnosed. Ultimately this is leading to an

increase in the development of resistance to drug treatments. Highlighting the need

for an improved system with a greater power of differentiating than the current T-

group genotyping system.


2.4.2 Genotyping based on cox1/2 sequences

Although rRNA genes prove the most attractive and popular choice for

intraspecific phylogeny and taxonomy studies, mitochondrial genes are also suitable

for use as a resolving tool to better understand close phylogenetic relationships

because they evolve approximately 10 fold faster than nuclear genes (Brown et al.,

1982). The mitochondrial genome is most likely to be under different constraints, as

it is located within a cell organelle rather than the nucleus (Ledee et al., 2003). The

Acanthamoeba mitochondrial genome appears to be conserved and have evolved

relatively slowly (Burger et al., 1995).

An earlier study has looked at sequence variation of cytochrome oxidase

genes within Acanthamoeba species, revealing a large degree of nucleotide

variation, enough to differentiate eight AK patient isolates, as well as matching six

of them by sequence homology to their respective tap water isolates (Kilvington et

al., 2004). Here the same mitochondrial gene cox1/2 was trialled as a suitable

contender for an alternative/additional target for genotyping Acanthamoeba species

rather than the current 18S T-group system.

The high degree of nucleotide variation within cox1/2 sequences of

Acanthamoeba isolates made the search for suitable primers difficult. The primers

were required to be Acanthamoeba specific, and suitable for use with as many

strains as possible. A second objective was to design a pair of primers that produced

a sequence product ideally <1,000 bp, but still include a region of high variability.

The resulting primer pair CoxA-486F and CoxA-1057R is sensitive and highly

effective, with the ability to work well with a range of unusual strains. Even

amplifying the cox1/2 gene of an atypical AK amoeba AK95/1153, where the 18S

Acanthamoeba specific primers JDP1F and JDP2R failed, and only the universal


primers succeeded (18sF: 18sR). Sequence analysis of the strain AK95/1153 has

provided surprising results with its cytochrome oxidase sequence matching

Acanthamoeba, but its 18S sharing 96% homology with F. aegytia. Further research

is needed to understand more of its genetics and evolution, but initial observations

have revealed some binucleate cysts, further linking it to the Flamella genus. If

confirmed as a Flamella sp., this would be first time the genus had been associated

with a case of human keratitis.

Evidence has been presented here to indicate that an alternative system that

has the ability to resolve Acanthamoeba isolates with pathogenic potential has been

developed. Genotyping by cox1/2 sequences does not group multiple pathogenic

strains into a few clades, as does the T-group system. However, the majority of

strains included in this study are AK causing isolates, the addition of environmental

strains would show their effect on the branching patterns, and in turn highlight the

robustness of the system.

2.4.3 Typing with concatenated gene sequences

Recent studies have shown multiple housekeeping genes can be

concatenated to form one super-gene alignment for building more robust

phylogenetic trees, which provide better discrimination power (Devulder et al.,

2005, Gadagkar et al., 2005, Kurtzman and Robnett, 2007).

This concatenated approach within the Acanthamoeba genus using the well

documented SSU 18S rDNA gene from the T group system, in combination with

another housekeeping gene, the mitochondrial cox1/2, has resulted in trees with


greater powers of discrimination. Which ultimately improve the ability to

phylogenetically classify the subgenus and improve the diagnosis of AK.

The phylogenetic trees based on concatenated data trees bears most

resemblance to the trees based on cox1/2 data, and increases the discrimination

power of the phylogenetic analyses within Acanthamoeba spp., something that has

not yet been achieved by the less discriminatory T-group system.

Results here do not contradict current opinions, now strongly biased towards

using a multigene approach to increase the resolution power of identifying species

(Devulder et al., 2005, Gadagkar et al., 2005, Kurtzman and Robnett, 2007).

However, although the concatenated tree has more discriminatory capabilities than

those with 18S data alone, little improvement is gained over using cox1/2 data

independently. The addition of even more genes in a concatenated dataset should be

tested to ascertain if further discriminatory power would be obtained. Further

analysis should also be considered using a larger data set than studied here.

2.4.4 Multiple alleles and mixed infections using 18S and cox1/2

Historically, evidence has suggested the presence of a single allele of the

18S gene occurring within Acanthamoeba, despite estimates suggesting that A.

castellanii Neff has several hundred copies of the gene. More recent evidence has

shown multiple alleles are actually present in at least seven isolates (Stothard et al.,

1998). For the presence of multiple alleles of 18S to be due to errors by taq

polymerase, the same allelic-specific mistakes would have to occur in two groups of

isolates, which is unlikely. However multiple alleles can occur as a result of

genetically different strains being present within the initial amoebae cultures.


The discovery of multiple alleles can be a benefit, and aid a better

understanding of the course of the infection.

Using a repeat series of isolates collected from a single patient, a second

study determined if drug resistance to propamidine had developed, or if the failure

of the patient to respond to treatment was due to a mixed infection (Ledee et al.,

1998). The presence of identical 18S multiple alleles in both drug-sensitive and

drug-resistant isolates (pre and post treatment), led to the conclusion that resistance

had developed throughout the cause of the treatment.

This study has shown the 18S gene sequence from two strains isolated from

Patient 6 (AK 6 and AK 6a), cluster independently on two branches, each in a

different clade, AK 6a in T4 and AK 6 in T3. However when analysed by cox1/2

sequence, both strains show sequence homology, and are found together in cox1/2

group c. Suggesting Patient 6 was infected with a strain of Acanthamoeba which has

multiple alleles of it’s 18S gene. Unlike here, the previous study which uncovered

multiple alleles of 18S in seven isolates, found where they do occur, both alleles fall

within the same 18S rDNA sequence type, which as a consequence is unlikely to

have a major impact on genotyping (Stothard et al., 1998).

Analyses of 18S sequences from repeat isolates of Patient one (AK 1 and AK 1a)

showed the presence of two sequence types (cox1/2 c and h), with AK 1 located in

T11, and AK 1a in T3. However, analysis of the cox1/2 gene from both isolates

shows a similar pattern, with two distinct sequence types. Evidence no longer points

to multiple alleles in either genes, but instead a mixed infection within the one

patient.

Evidence for the presence of multiple alleles second gene, cox1/2, was found

in this study. Six isolates recovered from a single patient (three), collected as repeat


samples, demonstrated four sequence types (cox1/2 a, e, g, and h), while they shared

sequence homology when analysed by 18S. Multiple alleles have also been

discovered in the Acanthamoeba lactic dehydrogenase gene (Ward and al-Abidin,

1988), but no others as yet have been identified. This is likely to change with the

imminent release of the complete high-quality sequencing of A. castellanii Neff

genome, which is currently in progress.

The use of two genes in relatively quick, and robust (in the case of 18S, and

yet to be proven in cox1/2) PCR can rapidly confirm or rule out the possibility of

mixed infections within a patient. Early detection of a mixed infection would

benefit the patient by improving prognosis, and reducing the risk of drug resistance

developing. In addition, a genotyping system capable of identifying multiple alleles

allows drug resistance to be tracked throughout the course of the treatment.

2.4.5 Conclusion

Speciation of the amoebae involved in acanthamebiasis may have

implications regarding epidemiology and the course of the infection, but is unlikely

to have a huge effect on the treatment, while the discovery of a mixed infection in

an individual may. The combined use of two systems to genotype can rapidly

determine the occurrence of a mixed infection, where a single system would

struggle.

Evidence here, supported previous studies (Stothard et al., 1998) identifying

multiple alleles of 18S occurring within Acanthamoeba, and additionally recognised

multiple alleles of the cox1/2 gene present within the genus. Tracking multiples

alleles of both genes and using them as epidemiological markers can track drug

regime resistance.


The T-group system is proving an invaluable tool within epidemiological

studies, and is continuing to evolve with the addition of new genotypes. However

the T-group system does not effectively resolve the isolates with pathogenic

potential, and clumps large numbers of strains within groups T3, T4 and T11. The

search for an efficient, fast and robust system to differentiate between virulent and

non-virulent strains has been an ongoing concern. This study has presented cox1/2

as a suitable contender to 18S for use in a phylogenetic system in the

epidemiological typing of Acanthamoeba, as it provides much more distinction

between the strains than the T-group system. However, the combined use of both

systems in this study led to the discovery of the mixed infection in Patient one. This

had the potential to have been interpreted as multiple alleles had either system been

used independently. Leading ultimately to the conclusion that the use of both

systems to genotype Acanthamoeba is synergistic.

Swedish GAE study 93

3 SWEDISH GAE STUDY

3.1 Introduction

Species of Acanthamoeba are one of the causative agents of granulomatous

amoebic encephalitis also known as GAE. It is a rare disease, with less than 200

cases caused by Acanthamoeba documented within the literature (Schuster and

Visvesvara, 2004). The skin and the olfactory neuroepithelium are the most likely

route of infection (Walochnik et al., 2008). Once inside the body, the amoebae most

likely gain access into the CSF by hematogenous, or by passing directly through the

neuroepithelium (Marciano-Cabral, 2003). Typically the infections occur in the

brain and/or the lungs of severely debilitated or immune suppressed individuals.

The infection presents with hemorrhagic necrotising lesions, generally found in the

cerebrum, cerebellum and brain stem, with symptoms including headaches, fever,

behavioural abnormalities and hemiparesis (Marciano-Cabral, 2003, Martinez and

Janitschke, 1985, Schuster and Visvesvara, 2004, Walochnik et al., 2008).

Diagnosis is almost always made post-mortem following brain tissue

biopsies, and indirect immunofluorescence (IIF) staining of tissue sections (Bloch

and Schuster, 2005, Schuster and Visvesvara, 2004). However pre-mortem

diagnosis of GAE has been reported within the literature. Achieved as a result of

positive Acanthamoeba-specific PCR of several biopsy tissue and fluid specimens,

including cerebrospinal fluid (CSF), bronchoalveolar lavage specimens (BAL),

skin, lung, and brain tissue (from the main lesion). All of which had been

Acanthamoeba culture negative when tested (Walochnik et al., 2008).

On a rare occasion, direct recovery of Acanthamoeba from CSF has been reported

(Callicott, 1968, Sharma et al., 1993, Singhal et al., 2001), however amoebae are

generally not found in the CSF.


PCR has proved to be an invaluable tool, used to diagnose Acanthamoeba

infections. The highly sensitive T-group typing system has also been used to

determine epidemiological association between a keratitis-causing strain of

Acanthamoeba, the patient, their contact lens storage case and their domestic water

supply (Ledee et al., 1996). An alternative system based on mt DNA profiles,

comparing a region of the cox1/2 gene, has also been used to link six patients and

their home tap water isolates (Kilvington et al., 2004).

In an unprecedented case our laboratory received repeat clinical and water

samples from three patients from a single ward within a Swedish hospital. The

patients were immune compromised paediatrics with severe debilitating illnesses.

Extraordinarily, they had all contracted fatal GAE whilst being treated on the same

intensive care unit (ICU) within the hospital. The sample set included repeat CSF

samples from all three patients and hospital water samples from four shower rooms,

a pool, and the main in-water supply to the ward. Such an outbreak of

Acanthamoeba GAE infections has never before been documented, as occurrences

of cases previously have been sporadic involving individuals in isolated cases

(Khan, 2006, Marciano-Cabral, 2003, Visvesvara et al., 2007b).

Research has shown it is possible to build a profile of pathogenicity using a

range of in vitro assays, by assessing physiological characteristics of the amoebae

such as ability: to tolerate higher temperatures and osmolarity; to produce

extracellular proteases (Khan et al., 2000, Sissons et al., 2006); to cause cytotoxic

effects on epithelial cells (Khan et al., 2002, Walochnik et al., 2000); to withstand

human complement (Ferrante and Rowan-Kelly, 1983, Pumidonming et al., 2011,

Toney and Marciano-Cabral, 1998); and show increased sensitivity particularly of


cysts to antimicrobials (Alizadeh et al., 2009, Elder et al., 1994, Lim et al., 2000,

Perrine et al., 1995, Schuster and Visvesvara, 1998).

3.1.1 Aims

To build pathogenicity and molecular profiles of the infecting GAE

Acanthamoeba strains isolated from the Swedish hospital. Strains will be identified

and typed by morphology and genotype techniques, using 18S sequence analysis to

determine T-group and mitochondrial cox1/2 sequence analysis. The pathogenic

potential of the isolates will be analysed by means of an array of in vitro tests,

including trophozoite and cyst sensitivities to a range of antimicrobials. Analysis of

the Swedish hospital samples will determine if the patients are infected with the

same strain of Acanthamoeba, and if that strain is present within the hospital water

system, thereby identifying the source of the infection.


3.2 Materials and Methods

3.2.1 Establishing amoebae cultures

To establish the presence of FLA within the samples, aliquots were initially

cultured by monoxenic methods (section 2.2.3). All seeded NNA plates were

individually sealed with parafilm, and groups of plates incubated together in plastic

bags, at 30C. Plates were observed at 24-hour intervals to check for growth.

Attempts were made to establish all recovered FLA in axenic culture (section 2.2.4),

occurring over a period of weeks to months.

Amoebae were identified using phase contrast microscopy (inverted)

(Olympus CKX41), with a camera (Olympus, CAMEDIA C5050), following the

dichotomous key for morphological taxonomy (Page, 1988).

Once maintained in axenic culture, strains were cryopreserved for future

work (2.2.5).

All samples were collected from a single ward of Swedish hospital, and

kindly donated by Dr Elisabet Holst, Department of Laboratory Medicine, Lund

University, Lund, Sweden. Patient and hospital strains with as much corresponding

information regarding sample origin as we have access to, is shown in Table 7.

Unfortunately access has not been granted to additional information regarding the

collection of samples, or clinical information.

In addition to Swedish hospital isolates reference strains were included in

the analysis; AK associated A. polyphaga Ros, soil originating A. castellanii (Neff,

50370); GAE associated A. culbertsoni (30171) and A. healyi; and a strain not

associated with disease A. polyphaga (30871).


3.2.2 Amoebae DNA extraction

Axenic Acanthamoeba cultures were harvested for genomic DNA extraction

by the UNSET method (detailed in section 2.2.6). If attempts to axenise strains had

failed, the cells were then maintained and propagated by the alternative slower

method of monoxenic culturing (section 2.2.3). Acanthamoeba DNA was collected

directly from the agar surface but at lower concentrations than if obtained from

axenic culture. The cells were harvested by repeated washing of the NNA surface

with 8 ml of ice cold dPBS and a pipette. Recovered cells were resuspended in 1 ml

of ice cold dPBS, and subjected to DNA extraction by the UNSET method: Carried

as described in section 2.2.6.1 but with the following modifications: 0.6 ml UNSET

buffer; 0.6 ml phenol-chloroform (1:1); 0.6 ml chloroform:isoamyl alcohol (24:1);

and resuspended in 20 μl of TE0.1 buffer containing RNase A (5 μl/ml).

3.2.2.i Direct DNA extraction from CSF

DNA was extracted directly from CSF samples. Using a technique modified

from the guanidine acetate method to purify target DNA following PCR (section

2.2.10.2).

Added to 25 µl of CSF, were 500 µl of 7 M guanidine acetate and 20 µl of

silica in nH2O (section 2.2.10.2); Guanidine acetate creates a hydrophobic

environment, encouraging the nucleic acids to bind to the silica. The solution was

then heated for five minutes at 55C, mixed briefly by vortex, and incubated on ice

for 10 minutes. The mixture was centrifuged at 10,000 x g for 10 seconds and the

supernatant removed. The DNA bound silica was washed three times in 80%

isopropanol (volumes of 500 µl, 180 µl and 180 µl, respectively), spinning at

10,000 x g for 10 seconds and all supernatant removed between washes. Ensuring


the eppendorf cap was opened, the DNA bound silica was incubated at ambient

room temperature in a fume hood, for approximately 10 minutes, to allow the silica

pellet to completely air-dry. Finally the DNA was dissolved in 20 µl of pre-warmed

nH2O, and incubated for 5 minutes at 55C (with occasional mixing). The silica was

removed by centrifuging at 10,000 x g for 1 minute, and the supernatant containing

the DNA was recovered in to a fresh eppendorf.

3.2.3 DNA manipulation

Basic manipulation of DNA was carried out with PCR techniques as

described in section 2.2.9, and purification of PCR products as in 2.2.10. Clone

libraries within pGEM®-T Easy cloning vectors were established as described in

2.2.11, production of ultra competent E. coli as in 2.2.12, transformation as in

2.2.13, and purification of propagated plasmids as in 2.2.15. Plasmids were

sequenced commercially as detailed in section 2.2.16. Sequenced results were

analysed and manipulated as described in section 2.2.17. Isolates and clone libraries

were preserved for future use by cryopreservation techniques described in section

2.2.5.

3.2.4 Temperature tolerance assay

The method of Khan and colleagues was used to assess the ability to tolerate

temperature (Khan et al., 2001). This method was carried out by inoculating axenic

Acanthamoeba trophozoites (2.2.4) on to E. coli seeded NNA plates (as described in

2.2.3) and incubating them at 32C and 37C for up to 96 hours. Viability and

growth of the amoebae was determined by microscopy, and identification of the

leading edge of trophozoite migration across the plate surface.


3.2.5 Osmotolerance assay

The osmotolerance of the amoebae was assayed using a method of Khan and

colleagues to determine pathogenic potential (Khan et al., 2001). The method was

carried out as for temperature tolerance, with the exception of replacing normal

NNA plates (2.2.4) with NNA supplemented with 1 M mannitol (0.25 osmolar).

Plates were incubated at 28C and 37C for up to 96 hours. Growth of the amoebae

was determined by identification of the leading edge of trophozoite migration by

phase contrast microscopy. Mannitol supplemented plates, inoculated with A.

culbertsoni (30171) served as controls.

3.2.6 Protease secretion

A modified method of Maciver was used for extracellular protease secretion

zymography (Edinburgh). This involved sodium dodecyl sulphate-polyacrylamide

gel electrophoresis (SDS-PAGE) gels containing bovine gelatine (0.1%) (Heussen

and Dowdle, 1980), used to determine the presence of proteases in the

Acanthamoeba culture medium of the strains. Gels were cast containing 2.5 ml

1.5M Tris pH8.8, 4.31 ml dH2O, 100 μl 10% SDS, 44 μl of 22.5% bovine gelatine,

3 ml 30% acrylamide mix 50 μl 10% APS and 20 μl TEMED (modifications from

(Heussen and Dowdle, 1980, Khan et al., 2000)). Acanthamoeba culture medium

(ACM) from pathogenic and non-pathogenic species (5 μl) was mixed with

electrophoresis sample loading buffer (1:1) (2 ml glycerol, 0.5 g SDS, 8 ml 0.5M

Tris, 0.2 ml 0.5% bromephnol blue), and then applied to the gels. Gels were run at

200 V (Mini-PROTEAN®, Tetra Cell, BIO-RAD), in 1X electrode (running) buffer

pH 8.3 (10X: 3.03 g Tris base, 14.1 g glycine, 1 g SDS per 100 ml of dH2O). After

electrophoresis gels were incubated in solution (2.5% Triton X–100, 50mM Tris-


HCl pH 7.0) for 60 minutes, and then overnight in developing buffer (50mM Tris-

HCl pH 7.0, 2mM CaCl2) at room temperature. Gels were then stained with rocking

for 60 minutes (0.1% Coomassie blue R-250, 40% methanol, 10% acetic acid),

before being rinsed in tap water, and incubated with rocking in destain solution

(10% acetic acid, 5% methanol) for 90 minutes.

Areas of gelatine digestion were visualised as non-staining regions of the

gel. In experimental controls fresh culture media was used, which had not been in

contact with Acanthamoeba isolates.

3.2.7 Complement fixing potential

The method of Pumidonming and colleagues was used to assess the

complement fixing potential of the isolates (Pumidonming et al., 2011). This

method involved harvesting axenic trophozoites by centrifugation at 500 x g for 1

minute. Cells then were washed twice with ¼ strength Ringer’s solution, quantified

using a modified Fuchs Rosenthal haemocytometer, and adjusted to concentrations

of 1 x 104 per ml with ¼ strength Ringer’s solution.

By pooling the sera of four randomly selected healthy-individuals, normal

human serum (NHS) was produced; Whole blood was collected, and allowed to

clot, undisturbed at ambient temperature. All clots were removed by centrifugation

at 1,500 x g for 10 minutes at 8˚C, and the sera removed using a 3ml Pasteur

pipette. All sera were stored at –20˚C, in 0.5ml aliquots, to avoid freeze-thaw cycles

(Invitrogen, 2007).

In a round-bottomed microtitre plate (VWR, Lutterworth, U.K.) 100 μl of

trophozoite suspension was added to 100 μl of NHS. Heat inactivated NHS (56˚C


for 30 minutes) served as a control (Toney and Marciano-Cabral, 1998). Assay

plates were maintained 28˚C for 1 hour, and the morphology of the trophozoites

following incubation at 0, 5, 10, 20 and 60 minute time points, was examined using

phase contrast microscopy (inverted) (equipped with a Canon EOS 60D camera).

Living trophozoites appeared bright and moving, while killed trophozoites

had destroyed cell membranes. Effect of compliment on the amoebae was also

determined by the formation of a pellet in the bottom of the well. Pellets were only

formed in wells where lysis of the amoebae had occurred, and where healthy

trophozoites remained.

3.2.8 Cytopathogenic potential

A modified method of Khan and colleagues was used to determine the

cytopathic potential of the Acanthamoeba isolates (Khan et al., 2000). The method

involved observing the degradation of epthelioid HeLa cell monolayer’s by

Acanthamoeba trophozoites. Where by HeLa cells were grown in 25 cm2 tissue

culture flasks, in 5 ml Dulbecco’s Modified Eagle Medium (DMEM) (with L-

glutamine) supplemented with 10% FBS (heat inactivated) and incubated in CO2

(5%) at 37˚C, until 80% confluent. Adhered cells were washed in 10 ml dPBS, to

remove all traces of growth media, and detached from the flask surface by the

addition of 2.5 ml of 1X trypsin-EDTA solution and incubation at 32˚C for 5

minutes. Detached cells in suspension were added to 32.5 ml of pre-warmed (37˚C)

DMEM, and 2 ml added to all wells of a 12 well microtitre plate (VWR). Again

incubated in 5% CO2 at 37˚C, until 80-100% confluent.


Acanthamoeba trophozoites were grown as described in 2.2.4, in 5 ml of

semi-defined culture media in 25 cm2 tissue culture flasks, to a confluency of >

80%. Cells were harvested, by centrifugation (500 x g for 1 minute) and washed in

¼ strength Ringer’s solution (three times), before being quantified.

1000 trophozoites in 30 μl were added to each well of the microtitre plate

containing a confluent monolayer of HeLa cell, and incubated in 5% CO2 at 37˚C.

Cytopathogenic effect by the Acanthamoeba was identified as areas of cleared HeLa

cells, and observed by microscopy up to 96 hours post-inoculation. Controls with

epitheloid cells incubated alone but with 30 μl of either ¼ strength Ringer’s solution

or culture medium (without Acanthamoeba) were maintained at the same conditions

for comparisons to be made against.

3.2.9 Antimicrobial sensitivity assays

3.2.9.1 Trophozoites

The methods of Kilvington and colleagues were used for the trophozoite

antimicrobial sensitivity assays (Elder et al., 1994, Hughes and Kilvington, 2001).

This involves trophozoites grown as described in 2.2.4, in 50 ml of culture media in

175 cm2 tissue culture flasks with filter caps, to a confluency of > 80%. Cells were

harvested and washed three times in dPBS-Tween (5% w/v Tween® 80),

centrifuged at 500 x g for 5 minutes between each wash. Finally quantified using a

modified Fuchs Rosenthal haemocytometer, and adjusted to concentrations of 2 x

104 per ml in culture medium.

To test trophozoite sensitivity against antimicrobials, stock solutions of the

compounds, Voriconazole (a gift from Kemprotec Ltd, Maltby, U.K.), Hexamidine


Diisethionate (Désomedire 0.1%- Bausch & Lomb, Kingston upon Thames, U.K.),

Polyhexamethylene biguanide (PHMB) (Cosmosil CQ- Arch U.K. Biocides Ltd,

Castleford, U.K.), Natamycin (Pimaricin- Molekula Ltd, Audenshaw, U.K.) and

Miltefosine were prepared at 1000 μg/ml in nH2O (with the exception of

Miltefosine diluted in 5% ethanol), and filter sterilised (0.2 μm). Antimicrobial

compound dilutions were obtained by two-fold serial dilution of 100 µl with 100 µl

of ¼ strength Ringer’s solution across a flat-bottomed microtitre plate. Calibrated

trophozoites (100 μl) were added to each well, and the plates covered and incubated

in air, at 32˚C, for 48 hours. Compounds were assayed in triplicate, with controls

containing ¼ strength Ringer’s solution only without antimicrobial compounds.

Following incubation, trophozoites were examined by inverted microscopy,

and compared to the controls. Comparison to the test wells allows the degree of

trophozoite growth or destruction to be assessed. The Minimum Trophozoite

Inhibitory Concentration (MTIC) and the Minimum Trophozoite Amoebacidal

Concentration (MTAC) of each test compound against each Acanthamoeba strain

was determined. MTIC is defined as 50% inhibition of trophozoite replication

compared to controls, and MTAC, where all trophozoites are rounded, non-motile

and floating or lysed (Elder et al., 1994, Hughes and Kilvington, 2001).

3.2.9.2 Cysts

The methods of Kilvington and colleagues were also used for the cyst

antimicrobial sensitivity assays (Elder et al., 1994, Hughes and Kilvington, 2001).

This involves Acanthamoeba trophozoites grown as described in 2.2.4, in 50 ml of

semi-defined culture media in 175 cm2 tissue culture flasks with filter caps, to a

confluency of > 80%. From late log-phase cultures, cysts were prepared in Neff’s


constant pH encystment medium (Hughes and Kilvington, 2001, Neff et al., 1964).

Harvested trophozoites were washed in Neff’s encystment medium, and centrifuged

at 1000 x g for 5 minutes, three times. From the pellet approximately 107

trophozoites were added to 100 ml of Neff’s medium in a 175 cm2 tissue culture

flasks with vent/close caps, and left to encyst in a shake incubator revolving at 100

rpm, at 32˚C, for seven days (Hughes and Kilvington, 2001). Following

microscopic evaluation to confirm >90% of cysts were mature; cells were detached

from the flask surface by gentle rubbing of the flask wall with a modified cyst

remover. Made by inserting a sterile polyester tipped applicator (Pur-Wraps)

(Puritan Medical products, Maine, USA) into the upper end (mouthpiece) of sterile

1 ml x 0.01 Sterilin micropipette (Barloworld Scientific Ltd, Stone, U.K.), before

gently heating the micropipette in a Bunsen flame to cause the polystyrene pipette

to soften, and allow the upper end to be manipulated with sterile forceps to have a

90˚ bend. Harvested cells were pelleted by centrifugation at 1000 x g for 5 minutes.

The cyst pellet was then washed three times in ¼ strength Ringer’s solution (2.2.3),

and centrifuged at 1000 x g for 5 minutes. Finally cells were stored in ¼ strength

Ringer’s solution, and stored at 4˚C for use within 14 days.

Cysts were quantified using a modified Fuchs Rosenthal haemocytometer,

and adjusted to concentrations of 2 x 104 per ml in ¼ strength Ringer’s solution.

Antimicrobial cyst assays rely on the adherence of cysts to the bottom of the

polycarbonate wells of the microtitre plates, even after exposure to antimicrobial

compound and removal by washing. Cyst assays were carried out as for

trophozoites, until the end of the 48 hour incubation.


After which adhered cysts were washed three times in ¼ strength Ringer’s

solution with 15 minute soaking between rinses at room temperature. Wells were

then filled with 100 μl of ¼ strength Ringer’s solution containing E. coli JM101

(ATCC 33876) at an O.D.540 of 0.1-0.2, and incubated at 32˚C for 7 days.

Plates were examined for the presence of trophozoites by inverted

microscopy, and the Minimum Cystical Concentration (MCC) was recorded as the

lowest concentration of antimicrobial compound that resulted in no excystment and

trophozoite replication (Elder et al., 1994).


3.3 Results

3.3.1 Culture and morphological analysis

Diagnosis in suspected cases of GAE is isolation, cultivation and subsequent

microscopy of Acanthamoeba directly from the patient, followed by the sensitive

technique of 18S rDNA sequence analysis.

3.3.1.1 Patient one

Initial samples collected from the patient, tested culture positive for

Hartmannella sp. (Table 7). Hartmannella has yet to be found as the causative

agent of disease in humans, and so given the highly irregular occurrence of

culturing this species from a human sample, additional specimens from the patient

were requested. Of the subsequent samples, four were Acanthamoeba culture

positive (Table 7). The Hartmannella isolated from Patient one, could only be

maintained on seeded-NNA plates.

3.3.1.2 Patient two

The samples included five CSF specimens, collected over a period of five

months, all tested culture positive for Acanthamoeba sp. (Table 7). Axenic culture

was achieved but compared to the hospital water supply samples, growth remained

consistently slower.


3.3.1.3 Patient three

Of the two samples collected from Patient three, both proved culture positive

for Acanthamoeba sp. (Table 7). The earlier of the two specimens collected was a

sample of CSF, followed by a second sample taken from a head wound, two months

later. Again axenic culture was established but Patient three isolates remained

consistently slower when compared to the growth rate achieved by hospital water

supply samples.

3.3.1.4 Hospital water samples

Samples were collected from several areas within the ICU that the patients

had access to, and were associated with water, including a swimming pool, and

several shower rooms. The ICU pool and two shower rooms all tested culture

positive for Acanthamoeba. Axenic culture was established for all isolated strains.

Culture results obtained from this study were relayed back to the hospital,

which responded immediately to super chlorinate the water supply within the

hospital, in an attempt to rid the system of the FLA. Following this treatment,

subsequent samples were collected from the hospital, and subjected to prolonged

repeated analyses, despite continued attempts the sample proved to be culture

negative of FLA.

Repeat testing approximately one year later, identified the presence of

Acanthamoeba (BL2997) (Table 7).


Table 7. FLA isolated from clinical and hospital samples collected in a Swedish

hospital, with morphological identification of genus groups.

Source Strain Origin Culture Identification

Patient 1 a GAE-1 CSF + Hartmannella sp.

GAE- 1a ? - Na

GAE- 1b CSF - Na

GAE- 1c CSF - Na

GAE- 1d CSF - Na

GAE- 1e CSF - Na

GAE- 1f ? + Acanthamoeba sp.

GAE- 1g ? + Acanthamoeba sp.

GAE- 1h ? + Acanthamoeba sp.

GAE- 1i ? + Acanthamoeba sp.

Patient 2 a GAE- 2 CSF + Acanthamoeba sp.

GAE- 2a CSF + Acanthamoeba sp.

GAE- 2b CSF + Acanthamoeba sp.

GAE- 2c CSF + Acanthamoeba sp.

GAE- 2d CSF + Acanthamoeba sp.

Patient 3 a GAE- 3 CSF + Acanthamoeba sp.

GAE- 3a Head wound + Acanthamoeba sp.

ICU Pool a Pool Pool + Acanthamoeba sp.

Shower rooms a SR- 3 Number 3 + Acanthamoeba sp.

SR- 15 Number 15 + Acanthamoeba sp.

Post-super

chlorination a

BL2997 ICU in-water

supply

+ Acanthamoeba sp.


Na: Not applicable; +: Culture positive; -: Culture negative; ?: Source unknown; a:

Donated by Dr E. Holst, Lund University, Lund, Sweden.

3.3.2 Molecular analysis

Despite being culture negative, Acanthamoeba DNA was isolated from a

Patient one’s CSF sample (GAE- 1b). The DNA was directly extracted using a

method with guanidine acetate and silica.

Comparisons made between the PCR findings and those of the culture

methods (Table 8), showed 100% obtainment of 18S and cox1/2 sequence

fragments from all culture positive samples. While from the six culture negative

specimens, one PCR positive result was obtained.

3.3.2.1 18S phylogenetic analysis

A region of 18S rDNA was amplified, cloned and sequenced from all FLA

isolated from the three patients, and the hospital, this included all Acanthamoeba

strains and the single Hartmannella sp. from Patient one. Two 18S primer sets and

corresponding sequencing primers, were used in this study, to reproduce sufficient

sequence length (approximately 204 bp; 1,175-1,379 bp) to determine any variation

for T-group differentiation (Schroeder et al., 2001). All Acanthamoeba 18S rDNA

PCR’s were carried out using the primer pair JDP1F and JDP2R (Schroeder et al.,

2001), producing partial (450 bp) 18S rDNA products. For the Hartmannella sp., a

larger (631 bp) 18S product was obtained with the use of universal primer pair 18S

F & 18S R (Weekers et al., 1994).

Individual BLAST analysis was carried out on all sequences (Altschul et al.,

1997), with nucleotide blast parameters optimised for highly similar sequences


(megablast): Results show the partial 631 bp sequence from Patient one’s

Hartmannella sp. matched by 99% to H. vermiformis (GenBank AY680840

(Kudryavtsev et al., 2005)) .

BLAST analysis of Acanthamoeba sp. isolated from Patient one, Patient two

and Patient three, all matched 100% to Acanthamoeba sp. ATCC 50496 (GenBank

ASU07408 (Gast et al., 1996)). The only Acanthamoeba strain isolated from the

Swedish hospital with an 18S sequence that differed from the others was that of the

ICU pool: Its 18S rDNA sequence matched 100% with A. castellanii ATCC 30010

(GenBank EF554328.1 (Kohsler et al., 2008)).

Analysis of the 18S sequences of all repeat strains isolated from each

patient, confirmed that each was infected with only one strain of Acanthamoeba,

and therefore analysis from this point forwards, was carried out using only one

strain for each patient.

The 18S sequences for Acanthamoeba strains from the three patients and

hospital samples, were aligned to each other using the alignment program for DNA,

ClustalW (2.1) (Larkin M.A. et al., 2007), which allows single nucleotide

differences to be identified. ClustalW identified sequence variation between the two

Acanthamoeba sp. 18S sequence types from this study to be 1.2%: Corresponding to

eight differences in a length of sequence 227 bp long: Six of these were three

double bp omissions from the pool strain, and two were separately occurring base

replacements from cytosine (C) to thymine (T) again in the pool strain.

Using 18S sequences a neighbour-joining distance tree with Kimura two-

parameters correction for multiple substitutions were obtained using MEGA (5.05)

software. Included in the tree are sequences from Acanthamoeba reference strains:


A. castellanii (50373, Neff: type species), A. polyphaga (30871: not associated with

disease) A. polyphaga (Ros: associated with AK), A. healyi (CDC 1283:V013:

associated with GAE) and A. culbertsoni (30171, Lilly: associated with GAE),

along with the Swedish isolates from this study, and a sequence from one of the

closest genera to Acanthamoeba, H. vermiformis (Costa Rica; GenBank AY680840)

(Figure 7).

The tree (Figure 7) confirms the distinction between two strain types within

the Swedish hospital ICU water system: Patient and hospital strains with the

exception of the pool species are identical by 18S analysis; the pool strain differs by

1.5%. This tree confirms only one strain type caused the patients infection. With the

Swedish hospital isolates, from the three patients and the two shower rooms, clearly

cluster away from that of the differing Swedish isolate recovered from the ICU

pool. The pool strain is most similar to A. polyphaga Ros (with 1% dissimilarity).

However, all Swedish strains including the pool isolate belong to the T4 genotype,

based on dissimilarity values of greater than 5%.

The ICU pool strain is found on a branch of its own, supported by a high

bootstrap value of 95: This node clearly separates the T4 isolates (both A.

polyphaga strains and A. castellanii) and the other Swedish hospital strains. Most

species from within the T4 clade in this tree are less than 5% dissimilar from each

other: The only exception is the strain that has not been associated with disease A.

polyphaga (30871), which differs to the pool strain by more than 5% (the cut off

value found within a T-group genotype) (Stothard et al., 1998): However its

dissimilarity it less than the cut off value when compared to soil originating isolate

A. castellanii (Neff, ATCC 50370) and AK associated isolate A. polyphaga Ros.


A maximum parsimony (MP) tree was chosen over one analysed by

neighbour joining algorithms. As it more clearly, by associated bootstrap value

shows a distinction between the T4 strains and those associated with GAE (A.

culbertsoni and A. healyi). Bootstrap values of 99 are found at the node separating

GAE isolates from T4 and Swedish isolates. Unsurprisingly the branch containing

Hartmannella is clearly distinguished from that of Acanthamoeba, and here acts as

a root for the tree.

Patient 1 GAE- 1f

Hospital SR- 15

Hospital BL2997

Patient 3 GAE- 3

Patient 2 GAE- 2

Hospital SR- 3

A. polyphaga Ros

A. polyphaga 30871

A. castellanii 1501/1a

T4

Hospital Pool

T12 A. healyi


Patient 1 Hartmannella sp.

H. vermiformis

88

59

10

8

4

5

13

87

3

95

99


of Swedish hospital FLA with comparison species. The tree is unrooted and

obtained by Kimura two-parameters correction for multiple substitutions using

MEGA (5.05). Analysis is based on reference bp from 1,175 to 1,379. Designated

T-groups and strain origins are shown. Bootstrap values have been included, based

on 1,000 bootstrap values, and placed at the corresponding nodes.


3.3.2.2 Cox1/2 phylogenetic analysis

The BLAST analysis was repeated for the Swedish hospital strains, but this

time using an alternative sequence, that of the mitochondrial cox1/2 gene. Results

follow their 18S sequences, and confirm all Swedish strains except one, match each

other, and in the case of cox1/2, share 89% similarity with A. castellanii (GenBank

U12386.1). Again as with 18S sequence analysis, the differing strain was found to

be that of the ICU pool, which only shared 88% sequence homology with A.

castellanii (GenBank U12386.1).

Sequence variation between the aligned cox1/2 sequences of the ICU pool

strain, and each of those from the three patients and hospital water system, was 4%.

An MP tree was constructed with cox1/2 sequences (Figure 8), and included

the Swedish strains from this study, with the exception of the post-treatment strain

BL2997, and reference strains: A. castellanii (50373, Neff: type species), A.

polyphaga (30871: not associated with disease) A. polyphaga (Ros: associated with

AK), and A. culbertsoni (30171, Lilly: associated with GAE) (Figure 8). As with

18S analysis, two genotypes were represented here from samples collected from the

Swedish hospital ICU. When comparing MP analysis to NJ (Figure 9), the MP tree

contains three nodes with bootstrap values of over 90, while the NJ tree has three

nodes over 75, and one over 90. It is generally considered that bootstrap values of

greater than 70 are evidence to support the distinction of clades. Although the order

of strains is reversed between the two algorithms, the grouping of taxa remains the

same.


Phylogenetic analysis using cox1/2 compared to 18S sequence data provides

more differentiation between the species, where strains are not clumped together

into large genotypes (Figures 8 and 9).


T4 A. castellanii 1501/1a

Hospital Pool

A. polyphaga Ros

A. polyphaga 1501/3aT4

Hospital SR- 15

Patient 1 GAE- 1f

Patient 3 GAE- 3

Patient 2 GAE- 2

Hospital SR- 3

99

97

65

93

16

9

13






placed at the corresponding nodes.

Patient 3 GAE- 3

Hospital SR- 3

Hospital SR- 15

Patient 2 GAE- 2

Patient 1 GAE- 1f

A. polyphaga Ros

A. polyphaga 1501/3aT4

Hospital Pool

T4 A. castellanii 1501/1a


75

92

78

75

0.02







placed at the corresponding nodes. The scale bar represents the corrected number of

nucleotide substitutions per base using Kimura method.

3.3.3 Pathogenicity assays

3.3.3.1 Amoebicidal activity of human serum

Normal Human Serum (NHS) exhibited amoebicidal activity against

trophozoites of all seven strains of Acanthamoeba. Observations were made by

microscopy. Following 5 and 10-minute incubation trophozoites appeared live and

viable. After 20 minutes of incubation with the sera, cytopathogenic changes had

begun and trophozoites appeared distressed and rounded. A pellet of killed

Acanthamoeba had formed in all test wells, between 70-80 minutes of incubation

(Figure 10 A-C, and Table 8). Controls were as expected, with trophozoites

incubated for the same period of time in heat-inactivated NHS remaining viable.

A B C

Figure 10. Activity of Normal Human Serum against Acanthamoeba trophozoites.

A, effect of complement from NHS on Acanthamoeba trophozoites; B, control,


Acanthamoeba incubated with heat-inactivated NHS; C, pellet formed in the round

bottomed well after lysis of Acanthamoeba trophozoites.

3.3.3.2 Tolerance of increased temperature and osmolarity

All Acanthamoeba isolates were able to continue trophic growth on NNA E.

coli- seeded plates at both 32C and 37C. However, when challenged in conditions

of high sugar alcohol (1M mannitol), no strains of Acanthamoeba were able to

maintain trophic growth, at either temperature tested. Controls were as expected.

Cysticidal affect of high osmolarity (1M mannitol) was examined, with agar

slices cut from the incubated mannitol plates with trophozoites, and transferred onto

fresh NNA E. coli-seeded plates without mannitol (inverted). Trophic growth of all

strains resumed within 72 hours, and the leading edge was visible emerging from

the upturned slice. Although trophozoites were unable to survive in conditions of

high osmolarity, cysts were formed in the presence of the mannitol, and remained

viable and able to resume trophic growth once conditions became more favourable

(data not shown).

3.3.3.3 Protease secretion assays

Non-stained lanes of the SDS-PAGE confirm areas of gelatine digestion,

and extracellular protease activity. Zymography with ACM has shown all

Acanthamoeba assayed have the capacity to produce proteases (Figure 11). Lanes

associated with the Swedish isolates (1-3, 5 and 6) show a distinction from that of

AK causing A. polyphaga Ros (4). Controls were as expected, confirming proteases

are not present in sterile ACM.


Figure 11. Sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-

PAGE) containing gelatine, to observe the presence of extracellular proteases in

Acanthamoeba culture medium (ACM). Lane 1, ICU pool; 2, Shower room 3; 3,

Shower room 15; 4, A. polyphaga Ros; 5, Patient 3; 6, Patient 2; 7, Left blank; 8,

negative control, containing sterile culture media.

3.3.3.4 Cytopathogenicity of Acanthamoeba

In the cytopathogenicity assays, within 96 hours all Acanthamoeba species

had produced varying degrees of cytopathic effect on the HeLa cell monolayer

(Table 8). Of all the strains tested the non-pathogenic A. polyphaga 1501/3a (30871)

and Shower room 3 isolate, equally resulted in the lowest level of` HeLa cell

monolayer degradation. Obvious cleared areas within the monolayer were present,

containing trophic amoebae. Extensive degradation to the monolayer occurred in


wells containing the AK causing strain A. polyphaga Ros, again with numerous

trophic cells. While isolates derived from Patient two, ICU pool, and Shower room

15, all severely destroyed the HeLa cell monolayer with assays of each containing

an abundance of large amoebae full of vacuoles confirming digestion of food

particles. The strain isolated from Patient three caused the most epitheloid

monolayer degradation and resulted in the highest abundance of large digesting

trophozoites. Control assays were maintained for comparison, and both that with

sterile culture media and ¼ strength Ringer’s solution contained an intact fully

confluent HeLa cell monolayer.

Summaries of Acanthamoeba sp. pathogenicity characteristics are shown in

Table 8. Based on results obtained in this study, all strains that were tested are

shown to have at least the potential to be pathogenic. However, because the shower

room 3 isolate and A. polyphaga 1501/3a (30871) exhibited reduced rate of

cytopathic activity towards a monolayer of HeLa cells they could be considered less

pathogenic compared with the known Swedish GAE isolates and AK isolate A.

polyphaga Ros.


Table 8. Pathogenicity characteristics of Acanthamoeba sp. collected from an

outbreak of GAE with a single ward of a Swedish hospital, and compared with an

AK isolate and a non-pathogen.

Strain Growth Protease

secretion

Cytopathic

activity

Complement

lysis / min

T group

32C 37C 1M mannitol

Patient 2 ●● ●● - Yes +++ < 80 T4

Patient 3 ●● ●● - Yes +++ < 80 T4

Shower

room 3

●● ●● - Yes + < 80 T4

Shower

room 15

●● ●● - Yes +++ < 80 T4

ICU Pool ●● ● - Yes +++ < 80 T4

A.

polyphaga

Ros

●● ●● - Yes ++ < 80 T4

A.

polyphaga

1501/3a

(30871)

●● ●● - Yes + < 80 T4

●●: Good growth; ●: Growth; +: 0-25% of monolayer destroyed; ++: 25-50% of

monolayer destroyed; +++: 50-75% of monolayer destroyed. -: None.


3.3.4 Drug sensitivity assays

An array of antimicrobial compounds was tested on trophozoites and cysts

of the Acanthamoeba strains (Table 9 and 10).

Against trophozoites, sensitivity assays to determine the Minimum

Trophozoite Amoebicidal Concentration (MTAC) were completed (Table 9).

PHMB exhibited the best antimicrobial activity with average an MTAC of 2.9

μg/ml, followed by miltefosine with an MTAC of 11.7 μg/ml, and hexamidine with

an average MTAC of 23.4 μg/ml. Wide ranges of effective MTAC values were

observed with the therapeutic agents, natamycin and voriconazole. The average

MTAC of natamycin to exhibit antimicrobial activity was 187.5 μg/ml.

Voriconazole showed the highest MTAC, exhibiting antimicrobial activity between

a range of 3.9 - > 125 μg/ml. However the only strain to be killed by voriconazole

within the tested concentration range was AK isolate A. polyphaga Ros, with a

MTAC of 3.9 μg/ml, for all other strains against voriconazole, MTAC values were

greater than 125 μg/ml, and above the range tested (data not shown).

The minimum cysticidal concentration (MCC) was defined as the lowest

concentration of test compound that resulted in no excystment and trophozoite

replication after 48 hours of exposure. Against cysts miltefosine, voriconazole and

natamycin showed little activity, with MCC levels undetected at concentrations

greater than 500 μg/ml. Hexamidine and PHMB were the only therapeutic agents to

show antimicrobial activity towards cysts, within the concentration range tested.

Hexamidine resulted in antimicrobial activity towards 43% of the strains (3/7),

causing a detrimental effect to isolates from Patient 3, MCC level of 62.5 μg/ml, as

well as the pool and shower room 15 isolates, both with MCC levels of 1.9 μg/ml.


The compound to show the most activity against cysts was the disinfectant PHMB,

with MCC levels of 1.9 μg/ml for 100% of strains (Table 10).


Table 9. In vitro sensitivities of five antimicrobial compounds against several

strains of Acanthamoeba trophozoites and cysts.

MTAC μg/mla MCC μg/ml

a

Mean Range Mean Range

Miltefosine 11.7 7.8 - 15.6 >500 >500

Voriconazole >125 3.9 - >125 >500 >500

Natamycin 187.5 125 – 250 >500 >500

Hexamidine 23.4 15.6 – 31.25 >500 1.9 - >500

PHMB 2.9 1.9 – 3.9 1.9 1.9

MTAC: Minimum Trophozoite Amoebicidal Concentration; MCC: Minimum

Cysticidal Concentration; PHMB: Polyhexamethylene biguanide; a: n = 7.

Table 10. Acanthamoeba in vitro Minimum Cysticidal Concentrations (MCC),

against five antimicrobial compounds.

Strain MCC μg/ml

Miltefosine Voriconazole Natamycin Hexamidine PHMB

Patient 2 >500 >500 >500 >500 1.9

Patient 3 >500 >500 >500 62.5 1.9

Shower room

3

>500 >500 >500 >500 1.9

Shower room

15

>500 >500 >500 1.9 1.9

ICU Pool >500 >500 >500 1.9 1.9

A. polyphaga >500 >500 >500 >500 1.9


Ros

A. polyphaga

1501/3a

(30871)

>500 >500 >500 >500 1.9

PHMB: Polyhexamethylene biguanide


3.4 Discussion

3.4.1 Molecular

Both clinical and hospital samples were analysed for FLA using culturing

methods. Where upon recovery, all amoebae were morphologically classified to

genus level. All recovered Acanthamoeba strains were established axenic culture,

but despite continuous attempts the Hartmannella sp. could not be maintained this

way. Few publications in the literature suggests it is possible to maintain

Hartmannella vermiformis by axenic culture in modified PYNFH medium (Fields et

al., 1990): One group compared several Hartmannella strains in their study but only

H. vermiformis (ATCC 50237), was maintained by axenic techniques, the other two

were cultured in the presence of live, and heat-killed E. coli (Kuiper et al., 2006).

The use of culturing techniques obtained a 73% (16 of 22) recovery of

Acanthamoeba from the samples. In general it is relatively rare to recover viable

Acanthamoeba from CSF samples (Callicott, 1968, Sharma et al., 1993, Singhal et

al., 2001). More often, fluid specimens test PCR positive for Acanthamoeba, when

they have been culture negative. Here, one culture negative sample, tested positive

for Acanthamoeba by PCR using a direct DNA extraction method. With the use of

PCR techniques, 77% (17 of 22) of samples tested Acanthamoeba positive,

highlighting the sensitivity of PCR, and its ability to detect Acanthamoeba where

culturing methods failed. PCR requires the survival of DNA rather than the

numerous intact viable cells that must be present for culturing to be successful.

Although the 18S PCR is a sensitive technique (Ledee et al., 1996) because the gene

is present in such abundance within the cells, for further examination of the strains

and downstream assays such as pathogenicity tests, culturing of the isolates is

required.


Morphological classification of Acanthamoeba beyond genus level is

extremely difficult and can be subjective. The sensitive nature of PCR and the

ability to use it to classify strains into clades such as T-groups highlights the

valuable of molecular typing, and how it should be used in conjunction with original

culture and microscopic diagnostic techniques.

A previously designed PCR primer pair was used to amplify a highly

variable diagnostic region of the SSU 18S rDNA (204 bp) (Schroeder et al., 2001),

in addition to a pair designed for this study: Which amplifies a 564 bp segment of

the mitochondrial cox1/2 gene, modified from (Kilvington et al., 2004). Both genes

are attractive targets for species differentiation within Acanthamoeba.

The findings from molecular analysis carried out during this study show an

unprecedented case of three patients from the same ward within a Swedish hospital,

all being infected with a single strain of Acanthamoeba: this was confirmed by

identifying all isolates to have identical 18S gene sequences, as well as identical

cox1/2 sequences, by 18S could be typed into T4. A second strain was also

identified from the Swedish hospitals’ water system, originating from the pool: The

isolate was also found to belong to T4, but have different 18S and cox1/2 sequences,

from the others. Until now approximately 150 cases of GAE caused by

Acanthamoeba have been described worldwide, with T4 species as the predominate

group associated with both GAE and AK, followed by T1, T2, T5, T10 and T12

(Lackner et al., 2010, Walochnik et al., 2008).

A related amoeba from the genus Hartmannella was also isolated from the

Swedish hospital sample set, surprisingly originating from a patient. So far

Hartmannella has not reliably been reported as the cause of a fatal disease in

humans (Schuster and Visvesvara, 2004), but early reports within the literature


often referred to Acanthamoeba as Hartmannella, and used the genus names

interchangeably (Cleland et al., 1982, Jager and Stamm, 1972). Here, Hartmannella

has been associated with human disease, but most likely as an opportunistic

secondary species, as Acanthamoeba was also isolated from the same patient.

The interactions that occur between amoebae are a relatively unexplored

area, but could perhaps begin to explain why Hartmannella has only been found in

humans as a co-infection with Acanthamoeba. In the case of B. mandrillaris, cells

have been found to feed on Acanthamoeba sp. (personal communication, Dr W.

Heaselgrave). While a previous study identified a Hartmannella sp. in association

with a human infection, and determined it was unlikely to be the cause of the

infection, but rather an opportunistic coloniser which may have worsened the

disease (Centeno et al., 1996).

Free-living amoebae are increasingly being recovered from hospital systems,

unsurprising given the ubiquitous nature of Acanthamoeba (Carlesso et al., 2010,

Thomas et al., 2006): Most likely caused by increased frequency of testing and

improvements in techniques, arisen from a better understanding of the need to keep

opportunistic pathogens away from immune compromised hospital patients.

Microbial biodiversity within hospital water systems’ is an area of concern given

the association of FLA with amoeba-resisting bacteria (ARB) and viruses that have

the potential to cause infections in humans (Centeno et al., 1996). To rid a water

system of FLA is almost impossible, amoebae including Hartmannella sp. have

even been isolated from heavily cleaned hospital water systems (Thomas et al.,

2006). The Swedish hospital did however successfully rid their ICU water system of

Acanthamoeba by super chlorinating the system; unfortunately this was only

maintained for a limited period. Retesting approximately 1-year post treatment,


showed Acanthamoeba of the same genotype and cox1/2 sequence to be present

again.

3.4.2 Pathogenicity

Identification to a genus level is all that is required to diagnose

Acanthamebiasis. However by determining the pathogenicity of the isolate, and its

sensitivity to antimicrobial compounds, the potential to provide substantial benefit

to disease prognosis and the patient’s outcome is gained. Microscopy and

identification based on morphological characteristics alone are not sufficient to

determine the pathogenicity of Acanthamoeba sp.. However, studies have shown a

series of pathogenicity traits can be tested for, to distinguishing pathogenic from

non-pathogenic isolates (Khan et al., 2001, Khan et al., 2002, Khan et al., 2000).

Thermotolerance is an indicator of pathogenicity, (Khan et al., 2001, Khan

et al., 2002, Walochnik et al., 2000), and Acanthamoeba must be able to withstand

the internal temperature of the human body, at 37°C to cause GAE. All the Swedish

isolates were thermotolerant, and grew at 32°C and 37°C, although in the case of the

pool isolate its growth rate was slightly slower by comparison. Additionally both A.

polyphaga species were also thermotolerant, surviving not only 32°C but also 37°C.

Survival at 32°C is not surprising given that A. polyphaga (Ros) is an AK causing

species, originally isolated from a human cornea (Hughes and Kilvington, 2001).

However based on assays completed here, it appears to have the potential to survive

within a human body. Acanthamoeba polyphaga (30871) has previously been

shown to grow at 37°C, further supporting previous evidence to suggest that the

species should be considered pathogenic (Khan et al., 2002, Stothard et al., 1998).


Osmotolerance is another indicator of pathogenicity, and amoebae that have the

capacity to grow on agar with a higher osmolarity (1M mannitol) are more likely to

be pathogenic and be able to withstand the human body. Here only A. culbertsoni

(30171) the control, exhibited osmotolerance and grew on mannitol-containing

plates. Although all the Swedish isolates were able to form viable cysts that could

resume trophic growth once osmotic pressure had been removed. Previously,

evidence has shown A. polyphaga (30871) to exhibit osmotolerance (Khan et al.,

2002), yet here the strain was unable to grow on mannitol-containing plates. When

considering osmolarity in relation to pathogenicity characteristics, epidemiological

information for given isolates must be taken into consideration. Osmotolerant A.

griffini (1501/4) was classified as a weak pathogen, despite never having been

associated with disease, its ability to withstand increased osmotic environments is

more likely to attributed to having originated from the sea around the USA coast

(Khan et al., 2001). The exact reasons why these strains did not grow on mannitol-

NNA plates is unclear, but despite this all other tests suggest they were pathogenic.

To cause infection, Acanthamoeba have to infiltrate the human body’s

defence system. In ocular infections amoebae must cross the stromal layer of the

cornea, and the blood-brain barrier in encephalitis. Very little is known about the

exact mechanism of how the pathogenic cascade is achieved, but for invasion to

take place, interactions between the amoebae and host cells must occur, most likely

involving the host extracellular matrix (da Rocha-Azevedo et al., 2010). It has been

shown that secreted amoebic proteases have the ability to degrade components of

the host extracellular matrix, such as collagens I, III, and IV, elastin, fibronectin and

laminin (He et al., 1990, Hurt et al., 2003, Na et al., 2001, Sissons et al., 2006). By


inhibiting proteases, the capabilities of Acanthamoeba to invade collagen and BD

Matrigel (basement membrane matrix) are reduced (da Rocha-Azevedo et al.,

2010). Here all isolates including the Swedish water strains produced extracellular

proteases. The pathogenesis of Acanthamoeba infections is highly complex and

multifaceted involving proteolytic activity. Proteases are degradative enzymes that

are vital for migration and tissue invasion. All Acanthamoeba isolates tested to

date, including pathogenic and non-pathogenic produce extracellular proteases, to

catalyse the hydrolysis of large proteins into smaller molecules ready for absorption

by the cell (Khan, 2009).

Acanthamoeba are free-living organisms, which happen to be opportunistic

pathogens with a secondary ability of being able to cause infection. As all isolates

tested here produce extracellular proteases, it was not unexpected to observe that

they all too exhibit cytopathic effects on epitheloid cells when incubated together.

However, subtle differences were found between some of the strains, when

compared to each other. The length of time taken to reach the same point of

destruction varied amongst the strains: With isolates from shower room 3, and A.

polyphaga (30871) causing destruction to the HeLa cell monolayer at a slow rate,

which was only marginally slower than the AK isolate A. polyphaga (Ros).

The innate resistance factor complement is the host’s powerful first line of

defence against invading Acanthamoeba. Although many microorganisms are

susceptible to complement lysis, some are better suited to withstand its lytic activity

than others (Jokiranta et al., 1995), which includes the highly pathogenic species A.

culbertsoni (Toney and Marciano-Cabral, 1998). Resistance to complement by

some Acanthamoeba strains seems to show some correlation to pathogenicity,

although reports have shown both pathogenic and non-pathogenic strains are


susceptible to lysis by complement (Ferrante, 1991, Ferrante and Rowan-Kelly,

1983, Pumidonming et al., 2011).

Ultimately in vitro analyses have shown no single Swedish hospital strain to

be any greatly more virulent than any of the others, and in fact they have

comparable pathogenicity traits to A. polyphaga Ros and A. polyphaga 30871. For

an Acanthamoeba infection to take hold, multiple virulence factors must be met.

Which accounts for the rarity of the disease despite the ubiquitous nature of

Acanthamoeba within the environment. Never before has an outbreak of GAE been

documented, suggesting extraordinary circumstances had occurred within the

Swedish hospital involving and/or the amoebae or patients. Pathogenicity assays

have confirmed that all the Swedish hospital isolates are virulent but perhaps the

line between pathogenic and non-pathogenic is not so clearly defined, and because

all strains have the ability to produce extracellular proteases (Khan, 2009) for

example, they also have an innate capacity to cause disease, providing predisposing

factors allow.

To provide a fuller picture of virulence, several other pathogenicity assays

could have been included, such as longevity tests observing cyst endurance in

storage over an extended period of time. As well as alternative extracellular

proteases assays, identifying their presence: Invasion/migration assays can establish

if the trophozoites have the ability to pass through a layer of gelatine collagen, or

BD Matrigel (basement membrane matrix) (da Rocha-Azevedo et al., 2010, Hurt et

al., 2003) to a lower chamber containing nutrients. Proteases can be classified into

six major groups including, serine, aspartic, cysteine, metalloproteases, threonine,

and glutamic acid (Khan, 2009), determination of the type and also quantity

produced by each strain, will improve the understanding its pathogenicity potential.


A full examination of the Swedish hospital system should also be carried out,

identifying if the system supports large numbers/types of microorganisms making it

an ideal home to harbour pathogenic FLA.

3.4.3 Antimicrobial sensitivity

There is no current effective treatment for GAE or disseminated

acanthamebiasis, and very few people have ever responded favourably to therapy.

For antimicrobials to be successful in treating GAE, they like the amoebae must be

able to cross the blood-brain barrier, and enter the brain and CSF. Most

acanthamoeba infections, are treated with antimicrobial combinations, and of those,

success has been achieved with intravenous pentamidine isethionate in combination

with topical chlorhexidine and ketoconazole administered to a transplant patient

who contracted disseminated infection (Slater et al., 1994): An HIV/AIDS patient,

received fluconazole and sulfadiazine combined with surgical excision of the brain

lesion (Seijo Martinez et al., 2000): While a previously healthy paediatric, received

ketoconazole and had a total excision of the infected area (Ofori-Kwakye et al.,

1986): Combination therapy with oral miltefosine and amikacin were used to

successfully treat a patient with GAE in Austria (Walochnik et al., 2008). Another

successful outcome arose from a 17-year-old immunocompromised patient who

presented with acute purulent meningoencephalitis, and was treated with a

combination of meropenem, linezolid, moxifloacin, and fluconazole (Lackner et al.,

2010). The strategy of combining antimicrobial compounds may result in beneficial

synergistic effects, and additionally avoid any potential resistance patterns.

Particularly as in vitro sensitivity testing of clinical isolates results in strain and

species differences, highlighting that no single compound can be assumed effective


against all amoebae (Schuster and Visvesvara, 1998). Further compounded by the

poor correlation observed between in vitro MCC findings and patient response to

those therapeutic agents (Elder et al., 1994, Kilvington et al., 2002).

Unfortunately as access was not granted to any patient clinical information from

the Swedish outbreak, sensitivity assays were designed using a range of

antimicrobial compounds, with the intention of hopefully testing the strains in vitro

against a compound that they were exposed to during the course of treatment: If not

the actual compound, at least one belonging to the same family.

For this study in vitro cyst sensitivity testing, identified all strains could resist

cidal effects of miltefosine, voriconazole and natamycin at concentrations lower

than 500 μg/ml. Hexamidine also showed little effect towards cysts, at

concentrations lower than 500 μg/ml for four of the strains (Patient 2, Shower room

3, A. polyphaga Ros, and A. polyphaga 30871): Whilst being cysticidal to Patient

3’s isolate at concentrations above 62.5 μg/ml, and to Shower room 15 and the Pool

isolates at greater than 1.9 μg/ml. However good cysticidal activity against all

strains was detected in assays with PHMB (>1.9 μg/ml).

Polyhexamethylene biguanide (PHMB) is an antiseptic in the same family as

chlorhexidine but is not as cytotoxic. It is used to good effect as a treatment of AK,

where it is administered as a topical solution of 0.02%, either alone or in

combination therapy. As it has a low mammalian toxicity, it can therefore be

applied over long periods of time. Its structure is a positively charged polymer, with

low surface tension, that against Acanthamoeba targets the plasma membrane by

binding with the acidic phospholipids. As PHMB is a disinfectant it is only suitable

for topical application, and despite having good cysticidal effect on many strains of

Acanthamoeba it cannot be used to treat systemic infections. Several AK cases have


been documented, reporting no response to treatment with PHMB, even though

isolated amoebae were found to be sensitive to the compound in in vitro assays

(Elder et al., 1994, Murdoch et al., 1998, Perez-Santonja et al., 2003). Clearly

inconsistencies between in vitro sensitivity profiles and management of individual

cases of acanthamebiasis can occur. In vitro drug sensitivity assays are undoubtedly

important in the development of potential new compounds to treat acanthamoeba

infections, but such assays clearly have limitations: Can any effects occurring

within the carefully controlled environment of a microtitre well fully represent the

human body? Effects of the antimicrobials on the amoebae may be altered by host

cellular and humoral responses, potentially affecting the course of the disease.

This study reported some cysticidal effect caused by the antimicrobial agent,

hexamidine di-isethionate (Desomedine): The compound is a homologue of

propamidine but has been shown in some cases to have greater cysticidal activity

(Brasseur et al., 1994, Perrine et al., 1995), and disputed more recently (Seal, 2003).

The diamidine group, contains many homologues including not only propamidine

but, butamidine, heptamidine, hexamidine, octamidine, pentamidine and

nonamidine (Perrine et al., 1995). The antimicrobial properties of the diamidines’

against Acanthamoeba have been found to be proportional to the length of the alkyl

chain of the diamidine molecule, linked to its lipophilicity and ability to penetrate

the amoeba cell membrane. Which causes structural changes to the plasma

membrane leading to alterations of cell permeability (Perrine et al., 1995). Currently

the recommended treatment regime for AK includes an aromatic diamidine, either

hexamidine (Desomedine) or propamidine isethionate (Brolene) at 0.1% (1 mg/ml),


in combination with a disinfecting biguanide either 0.02% PHMB or chlorohexidine

digluconate (Khan, 2009).

Here, hexamidine against cysts showed variable amoebicidal results: with very

weak cytotoxic effects against four strains >500 μg/ml; limited towards another at

62.5 μg/ml, with strong activity shown against two strains, at 1.9 μg/ml. Variability

was also shown against trophozoites, with an average MTAC of 23.4 μg/ml, with a

range of 15.6 – 31.25 μg/ml. Such variability in the effectiveness of one agent

reflects the suggestion that natural variation in Acanthamoeba species accounts for

much of the variation found in sensitivity assays to determine cidal effects caused

by antimicrobials (Elder et al., 1994, Khan, 2009).

In all cases by in vitro assays, trophozoites were more susceptible to

therapeutic compounds than cysts, as has been previously reported (Elder et al.,

1994, Jones et al., 1975, Walochnik et al., 2009). As with cysts assays, PHMB

showed the most activity against strains, with average MTAC values of 1.9 µg/ml.

Next best amoebicidal activity towards the strains was shown by the promising

compound Miltefosine (Impavido®, hexadecylphosphochlorine): An

alkylphosphocholine anticancer agent that appears to hold necessary therapeutic

properties for the treatment of AK (Walochnik et al., 2009), and GAE (Walochnik

et al., 2008). Although the mode of action is unknown, miltefosine appears to

accumulate in the cell membrane causing structural disruption and affecting cell

metabolism (Eibl and Unger, 1990, Walochnik et al., 2009). Recently it has been

used for the treatment of disseminated Acanthamoeba infections, and is also a

potential antimicrobial agent for treatment of AK and GAE (Mrva et al., 2011).

Miltefosine has been shown to be effective against clinical and environmental


isolates of Acanthamoeba, starting at concentrations of 2 µg/ml (~5 μM) (Schuster

et al., 2006a, Walochnik et al., 2002), and 25.5 μg/ml (62.5 μM) (Mrva, M 2011):

Here assays showed amoebicidal effects at 11.7 μg/ml, between the strong and weak

amoebicidal activity previously shown.

Natamycin (Natacyn®, 5% (50 mg/ml)) is a macrolide polyene antifungal

agent used to treat fungal eye infections. The macrolide antibiotic group is large,

containing many agents including amphotericin B, erythromycin, azithromycin and

clarithromycin. The macrolides have broad-spectrum activities against a range of

microorganisms (Mattana et al., 2004), and have been used to treat GAE, with both

successful outcomes (Nachega et al., 2005, Walia et al., 2007), and failure (Kuashal

et al., 2008). The precise mode of action of the macrolides within Acanthamoeba is

not fully understood but thought to involve the plasma membrane and ergosterol

(Raederstorff and Rohmer, 1985), altering cell permeability (Mattana et al., 2004).

Antifungal agents, natamycin (>500 μg/ml), and voriconazole (>500 μg/ml)

showed only weak amoebicidal effects against cysts, but higher cytotoxicity against

trophozoites was attained, natamycin (187.5 μg/ml) and voriconazole (>125 μg/ml).

Voriconazole belongs to the azole group, and includes the homolog compounds,

clotrimazole, miconazole, ketoconazole, and fluconazole. It has a strong inhibitory

effect upon Acanthamoeba trophozoites but not cysts (Schuster et al., 2006a), with a

mode of action that is similar to that of natamycin (Sanati et al., 1997).

Through the use of in vitro sensitivity assays, several pharmaceutical agents

have been shown to have antimicrobial properties towards Acanthamoeba cysts and

trophozoites, but no single compound has proven effective against all isolates of the

genus: Reflecting the high natural variability within the species. This only goes to


highlight the importance of carrying out drug sensitivity tests for every case of AK

and GAE diagnosed. Which would not only help identify the best choice compound,

but also help discount any that are potentially ineffective. The Amoebae Laboratory,

at the University of Leicester now routinely carries out such sensitivity testing for

the Leicester Royal Infirmary, and other UK hospitals. Their experience has shown

clinical strains of AK usually show sensitivity to biguanides, but sensitivity to all

other drug compounds in particular diamidines, is very variable (personal

communication Dr W. Heaselgrave). So it is important to identify the effective

treatments quickly to improve patient prognosis. However, cases that fail to respond

to medical therapy do occur despite the apparent in vitro sensitivity to compounds

such as PHMB (Elder et al., 1994, Murdoch et al., 1998, Perez-Santonja et al.,

2003). The cause of this is unknown, but for these patients prognosis is poor, as they

remain culture positive and have to undergo repeated penetrating keratoplasty. With

Acanthamoeba infections, prompt instigation of treatment is necessary for good

prognosis. Therefore sensitivity testing is important to ensure therapy stands a

chance of being effective and is not a waste of time.

3.4.1 Conclusion

Epidemiological implications and a better understanding of the course of

infection may occur as a result of speciation of the isolate involved in

acanthamebiasis. The value of the sequencing systems has been clearly shown here,

by allowing strain differentiation. Severely debilitated patients on a paediatric ICU,

within a Swedish hospital fell ill with symptoms of GAE. Clinical samples were

collected from the three patients and any water-associated places that the patients


had access to. Epidemiological typing was carried out, with CSF samples cultured

for FLA, confirming the patients were all infected with T4 Acanthamoeba isolates.

In addition to the Acanthamoeba, one patient had also contracted an extremely

rare associated infection with a species of Hartmannella as well.

The combined use of two molecular genotyping systems both 18S and cox1/2,

confirmed the infection had originated from the water system within the hospital

and the patients were all infected with the same strain. The T-group and cox1/2

systems both proved to be independently rapid, sensitive and highly effective tools

for use within epidemiological studies.

Pathogenicity and antimicrobial sensitivity assays confirmed the isolates

originating from the Swedish hospital were all virulent, but no more pathogenic

than expected given that they were associated with an unprecedented outbreak of

GAE.

References 138

4 REFERENCES

Adam, K. M. G., Blewett, D. A. & Flamm, W. G. 1969. The DNA of

Acanthamoeba spp; a method for extraction and its characterization. Journal

of Protozoology, 16, 6-12.

Adekambi, T., Ben Salah, S., Khlif, M., Raoult, D. & Drancourt, M. 2006. Survival

of environmental mycobacteria in Acanthamoeba polyphaga. Applied and

Environmental Microbiology, 72, 5974-5981.

Adl, S. M., Simpson, A. G., Farmer, M. A., Andersen, R. A., Anderson, O. R.,

Barta, J. R., Bowser, S. S., Brugerolle, G., Fensome, R. A., Fredericq, S.,

James, T. Y., Karpov, S., Kugrens, P., Krug, J., Lane, C. E., Lewis, L. A.,

Lodge, J., Lynn, D. H., Mann, D. G., Mccourt, R. M., Mendoza, L.,

Moestrup, O., Mozley-Standridge, S. E., Nerad, T. A., Shearer, C. A.,

Smirnov, A. V., Spiegel, F. W. & Taylor, M. F. 2005. The new higher level

classification of eukaryotes with emphasis on the taxonomy of protists.

Journal of Eukaryotic Microbiology, 52, 399-451.

Aksozek, A., Mcclellan, K., Howard, K., Niederkorn, J. Y. & Alizadeh, H. 2002.

Resistance of Acanthamoeba castellanii cysts to physical, chemical, and

radiological conditions. Journal of Parasitology, 88, 621-623.

Alizadeh, H., Neelam, S. & Cavanagh, H. D. 2009. Amoebicidal activities of

alexidine against 3 pathogenic strains of acanthamoeba. Eye and Contact

Lens, 35, 1-5.

Allen, R. D. 1961. A new theory of ameboid movement and protoplasmic

streaming. Experimental Cell Research, Suppl 8, 17-31.

Allen, R. D., Francis, D. W. & Nakajima, H. 1965. Cyclic Birefringence Changes in

Pseudopods of Chaos Carolinensis Revealing the Localization of the Motive

Force in Pseudopod Extension. Proceedings of the National Academy of

Sciences, 54, 1153-1161.

Alsam, S., Jeong, S. R., Sissons, J., Dudley, R., Kim, K. S. & Khan, N. A. 2006.

Escherichia coli interactions with Acanthamoeba: a symbiosis with

environmental and clinical implications. Journal of Medical Microbiology,

55, 689-694.

Altschul, S. F., Madden, T. L., Schaffer, A. A., Zhang, J., Zhang, Z., Miller, W. &

Lipman, D. J. 1997. Gapped BLAST and PSI-BLAST: a new generation of

protein database search programs. Nucleic Acids Research, 25, 3389-3402.

Axelsson-Olsson, D., Waldenstrom, J., Broman, T., Olsen, B. & Holmberg, M.

2005. Protozoan Acanthamoeba polyphaga as a potential reservoir for

Campylobacter jejuni. Applied and Environmental Microbiology, 71, 987-

992.

Barker, J. & Brown, M. R. W. 1994. Trojan Horses of the microbial world: protozoa

and the survival of bacterial pathogens in the environment. Microbiology,

140, 1253-1259.

Barker, J., Humphrey, T. J. & Brown, M. W. 1999. Survival of Escherichia coli

O157 in a soil protozoan: implications for disease. FEMS Microbiology

Letters, 173, 291-295.

Bharathi, M. J., Ramakrishnan, R., Meenakshi, R., Mittal, S., Shivakumar, C. &

Srinivasan, M. 2006. Microbiological diagnosis of infective keratitis:

comparative evaluation of direct microscopy and culture results. British

Journal of Ophthalmology, 90, 1271-1276.

References 139

Biddick, C. J., Rogers, L. H. & Brown, T. J. 1984. Viability of pathogenic and

nonpathogenic free-living amoebae in long-term storage at a range of

temperatures. Applied and Environmental Microbiology, 48, 859-860.

Blanton, W. E. & Villemez, C. L. 1978. Molecular Size and Chain Length

Distribution in Acanthamoeba Cellulose*. Journal of Eukaryotic

Microbiology, 25, 264-267.

Bloch, K. C. & Schuster, F. L. 2005. Inability to make a premortem diagnosis of

Acanthamoeba species infection in a patient with fatal granulomatous

amebic encephalitis. Journal of Clinical Microbiology, 43, 3003-3006.

Boom, R., Sol, C. J., Salimans, M. M., Jansen, C. L., Wertheim-Van Dillen, P. M.

& Van Der Noordaa, J. 1990. Rapid and simple method for purification of

nucleic acids. Journal of Clinical Microbiology, 28, 495-503.

Booton, G. C., Kelly, D. J., Chu, Y. W., Seal, D. V., Houang, E., Lam, D. S., Byers,

T. J. & Fuerst, P. A. 2002. 18S ribosomal DNA typing and tracking of

Acanthamoeba species isolates from corneal scrape specimens, contact

lenses, lens cases, and home water supplies of Acanthamoeba keratitis

patients in Hong Kong. Journal of Clinical Microbiology, 40, 1621-1625.

Booton, G. C., Visvesvara, G. S., Byers, T. J., Kelly, D. J. & Fuerst, P. A. 2005.

Identification and distribution of Acanthamoeba species genotypes

associated with nonkeratitis infections. Journal of Clinical Microbiology,

43, 1689-1693.

Brasseur, G., Favennec, L., Perrine, D., Chenu, J. P. & Brasseur, P. 1994.

Successful treatment of Acanthamoeba keratitis by hexamidine. Cornea, 13,

459-462.

Brown, W. M., Prager, E. M., Wang, A. & Wilson, A. C. 1982. Mitochondrial DNA

sequences of primates: tempo and mode of evolution. Journal of Molecular

Evolution, 18, 225-39.

Brzeska, H., Young, R., Knaus, U. & Korn, E. D. 1999. Myosin I heavy chain

kinase: cloning of the full-length gene and acidic lipid-dependent activation

by Rac and Cdc42. Proceedings of the National Academy of Sciences of the

United States of America, 96, 394-9.

Burger, G., Plante, I., Lonergan, K. M. & Gray, M. W. 1995. The mitochondrial

DNA of the amoeboid protozoon, Acanthamoeba castellanii: complete

sequence, gene content and genome organization. Journal of Molecular

Biology, 245, 522-537.

Burgos, J. M., Begher, S., Silva, H. M., Bisio, M., Duffy, T., Levin, M. J., Macedo,

A. M. & Schijman, A. G. 2008. Molecular identification of Trypanosoma

cruzi I tropism for central nervous system in Chagas reactivation due to

AIDS. American Journal of Tropical Medicine and Hygiene, 78, 294-7.

Byers, T. J. 1986. Molecular biology of DNA in Acanthamoeba, Amoeba,

Entamoeba, and Naegleria. International Review of Cytology, 99, 311-341.

Byers, T. J., Hugo, E. R. & Stewart, V. J. 1990. Genes of Acanthamoeba:

DNA,RNA and Protein Sequences (A Review). Journal of Protozoology, 37,

17S-25S.

Byers, T. J., Rudick, V. L. & Rudick, M. J. 1969. Cell size, macromolecule

composition, nuclear number, oxygen consumption and cyst formation

during two growth phases in unagitated cultures of Acanthamoeba

castellanii. Journal of Protozoology, 16, 693-9.

References 140

Callicott, J. H., Jr. 1968. Amebic meningoencephalitis due to free-living amebas of

the Hartmannella (Acanthamoeba)-Naegleria group. American Journal of

Clinical Pathology, 49, 84-91.

Carlesso, A. M., Artuso, G. L., Caumo, K. & Rott, M. B. 2010. Potentially

pathogenic Acanthamoeba isolated from a hospital in Brazil. Current


Castellani, A. 1930 An amoeba found in cultures of a yeast. Preliminary note. .

Journal of Tropical Medicine and Hygiene, 33, 160.

Cavalier-Smith, T. 1993. Kingdom protozoa and its 18 phyla. Microbiological

Reviews, 57, 953-94.

Cavalier-Smith, T. 1998. A revised six-kingdom system of life. Biological Reviews

of the Cambridge Philosophical Society, 73, 203-66.

Centeno, M., Rivera, F., Cerva, L., Tsutsumi, V., Gallegos, E., Calderon, A., Ortiz,

R., Bonilla, P., Ramirez, E. & Suarez, G. 1996. Hartmannella vermiformis

isolated from the cerebrospinal fluid of a young male patient with

meningoencephalitis and bronchopneumonia. Archives of Medical Research,

27, 579-86.

Cerva, L. 1981. Spontaneous occurrence of antibodies against pathogenic amoebae

of the limax group in domestic animals. Folia parasitologica, 28, 105-108.

Cerva, L., Serbus, C. & Skocil, V. 1973. Isolation of limax amoebae from the nasal

mucosa of man. Folia parasitologica, 20, 97-103.

Chappell, C. L., Wright, J. A., Coletta, M. & Newsome, A. L. 2001. Standardized

method of measuring Acanthamoeba antibodies in sera from healthy human

subjects. Clinical and Diagnotic Laboratory Immunology, 8, 724-30.

Chung, D. I., Yu, H. S., Hwang, M. Y., Kim, T. H., Kim, T. O., Yun, H. C. & Kong,

H. H. 1998. Subgenus classification of Acanthamoeba by riboprinting.

Korean Journal of Parasitology, 36, 69-80.

Cirillo, J. D., Falkow, S., Tompkins, L. S. & Bermudez, L. E. 1997. Interaction of

Mycobacterium avium with environmental amoebae enhances virulence.

Infection and Immunity, 65, 3759-3767.

Clarke, D. W. & Niederkorn, J. Y. 2006. The pathophysiology of Acanthamoeba

keratitis. Trends in Parasitology, 22, 175-180.

Cleland, P. G., Lawande, R. V., Onyemelukwe, G. & Whittle, H. C. 1982. Chronic

amebic meningoencephalitis. Archives of Neurology, 39, 56-57.

Cole, K. D. 1991. Purification of plasmid and high molecular mass DNA using

PEG-salt two-phase extraction. BioTechniques, 11.

Corsaro, D. & Venditti, D. 2010. Phylogenetic evidence for a new genotype of

Acanthamoeba (Amoebozoa, Acanthamoebida). Parasitology Research,

107, 233-8.

Culbertson, C. G. 1971. The pathogenicity of soil amebas. Annual Review of


Culbertson, C. G. & Harper, K. 1984. Pathogenic free-living amebae.

Immunocytologic demonstration and species identification. American

Journal of Tropical Medicine and Hygiene, 33, 851-6.

Culbertson, C. G., Smith, J. W. & Minner, J. R. 1958. Acanthamoeba: Observations

on Animal Pathogenicity. Science, 127, 1506.

Cursons, R. T., Brown, T. J., Keys, E. A., Moriarty, K. M. & Till, D. 1980.

Immunity to pathogenic free-living amoebae: role of humoral antibody.

Infection and Immunity, 29, 401-407.

References 141

Da Rocha-Azevedo, B., Jamerson, M., Cabral, G. A. & Marciano-Cabral, F. 2010.

Acanthamoeba culbersoni: Analysis of amoebic adhesion and invasion on

extracellular matrix components collagen I and laminin-I. Experimental

Parasitology, 126, 79-84.

Da Rocha-Azevedo, B., Tanowitz, H. B. & Marciano-Cabral, F. 2009. Diagnosis of

infections caused by pathogenic free-living amoebae. Interdisciplinary

Perspectives on Infectious Diseases, 2009, 251406.

Daggett, P. M., Lipscomb, D., Sawyer, T. K. & Nerad, T. A. 1985. A molecular

approach to the phylogeny of Acanthamoeba. Biosystems, 18, 399-405.

Dart, J. K., Saw, V. P. & Kilvington, S. 2009. Acanthamoeba keratitis: diagnosis

and treatment update 2009. American Journal of Ophthalmology, 148, 487-

499.e2.

De Jonckheere, J. F. 1980. Growth characteristics, cytopathic effect in cell culture,

and virulence in mice of 36 type strains belonging to 19 different

Acanthamoeba spp. Applied and Environmental Microbiology, 39, 681-685.

De Jonckheere, J. F. 1983. Isoenzyme and Total Protein Analysis by Agarose

Isoelectric Focusing and Taxonomy of the Genus Acanthamoeba. Journal of

Protozoology, 30, 701-706.

De Jonckheere, J. F. 2004. Molecular definition and the ubiquity of species in the

genus Naegleria. Protist, 155, 89-103.

Devulder, G., Perouse De Montclos, M. & Flandrois, J. P. 2005. A multigen

approach to phylogenetic analysis using the genus Mycobacterium as a

model. International Journal of Systematic and Evolutionary Microbiology,

55, 293-302.

Douglas, M. 1930. Notes on the classification of the amoeba found by Castellani in

cultures of a yeast-like fungus. Journal of Tropical Medicine and Hygiene,

33.

Dubey, J. P., Benson, J. E., Blakeley, K. T., Booton, G. C. & Visvesvara, G. S.

2005. Disseminated Acanthamoeba sp. infection in a dog. Veterinary


Dudley, R., Matin, A., Alsam, S., Sissons, J., Maghsood, A. H. & Khan, N. A. 2005.

Acanthamoeba isolates belonging to T1, T2, T3, T4 but not T7 encyst in

response to increased osmolarity and cysts do not bind to human corneal

epithelial cells. Acta Tropica, 95, 100-8.

Dykova, I. & Lom, J. 2004. Advances in the knowledge of amphizoic amoebae

infecting fish. Folia Parasitologica, 51, 81-97.

Edinburgh, T. U. O. Detection of Amoebal Proteases [Online]. Available:

http://www.bms.ed.ac.uk/research/others/smaciver/amoebal_proteases.htm

[Accessed 21/11/2011].

Eibl, H. & Unger, C. 1990. Hexadecylphosphocholine: a new selective antitumor

drug. Cancer treatment Reviews, 17, 2061-2074.

Eichinger, L., Lee, S. S. & Schleicher, M. 1999. Dictyostelium as model system for

studies of the actin cytoskeleton by molecular genetics. Microscopy

Research and Technique, 47, 124-34.

Ekelund, F. & Ronn, R. 1994. Notes on protozoa in agricultural soil with emphasis

on heterotrophic flagellates and naked amoebae and their ecology. FEMS

Microbiology Reviews, 15, 321-353.

Elder, M. J., Kilvington, S. & Dart, J. K. G. 1994. A clinicopathological study of in-

vitro sensitivity testing and Acanthamoeba Keratitis Investigative

Ophthalmology and Visual Science, 35, 1059-1064.

http://www.bms.ed.ac.uk/research/others/smaciver/amoebal_proteases.htm

References 142

Ferrante, A. 1991. Immunity to Acanthamoeba. Reviews of Infectious Diseases, 13,

S403-S409.

Ferrante, A. & Rowan-Kelly, B. 1983. Activation of the alternative pathway of

complement by Acanthamoeba culbertsoni. Clinical and Experimental

Imunnology, 54, 477-485.

Fields, B. S., Nerad, T. A., Sawyer, T. K., King, H., Barbaree, J. M., Martin, W. T.,

Morrill, W. E. & Sanden, G. N. 1990. Characterisation of an axenic strain of

Hartmannella vermiformis obtained from an investigation of nosocomial

legionellosis. Journal of Protozoology, 37, 581-583.

Gadagkar, S. R., Rosenberg, M. S. & Kumar, S. 2005. Inferring species phylogenies

from multiple genes: Concatenated sequence tree versus consensus tree.

Journal of Experimental Zoology Part B: Molecular and Developmental

Evolution, 304B, 64-74.

Garate, M., Cao, Z., Bateman, E. & Panjwani, N. 2004. Cloning and

characterization of a novel mannose-binding protein of Acanthamoeba.

Journal of Biological Chemistry, 279, 29849-56.

Gast, R. J. 2001. Development of an Acanthamoeba-specific reverse dot-blot and

the discovery of a new ribotype. Journal of Eukaryotic Microbiology, 48,

609-615.

Gast, R. J., Fuerst, P. A. & Byers, T. J. 1994. Discovery of group I introns in the

nuclear small subunit ribosomal RNA genes of Acanthamoeba. Nucleic

Acids Research, 22, 592-596.

Gast, R. J., Ledee, D. R., Fuerst, P. A. & Byers, T. J. 1996. Subgenus systematics of

Acanthamoeba: four nuclear 18S rDNA sequence types. Journal of

Eukaryotic Microbiology, 43, 498-504.

Gast, R. J., Sanders, R. W. & Caron, D. A. 2009. Ecological strategies of protists

and their symbiotic relationships with prokaryotic microbes. Trends in


Gelman, B. B., Rauf, S. J., Nader, R., Popov, V., Borkowski, J., Chaljub, G., Nauta,

H. W. & Visvesvara, G. S. 2001. Amoebic encephalitis due to Sappinia

diploidea. Journal of the American Medical Association, 285, 2450-2451.

Giegerich, R., Meyer, F., Schleiermacher, C, 1996. GeneFisher-Software Support

for the Dection of Postulated Genes. Proceedings of the Fourth

International Conference on Intelligent Systems for Molecular Biology, 68-

77.

Greub, G., La Scola, B. & Raoult, D. 2004. Amoeba-resisting Bacteria Isolated

from Human Nasal Swabs by Amoebal Coculture. Emerging Infectious

Diseases, 10, 470-477.

Greub, G. & Raoult, D. 2004. Microorganisms resistant to free-living amoebae.

Clinical Microbiology Reviews, 17, 413-433.

Guarner, J., Bartlett, J., Shieh, W. J., Paddock, C. D., Visvesvara, G. S. & Zaki, S.

R. 2007. Histopathologic spectrum and immunohistochemical diagnosis of

amebic meningoencephalitis. Modern Pathology, 20, 1230-7.

Gunderson, J. H. & Sogin, M. L. 1986. Length variation in eukaryotic rRNAs: small

subunit rRNAs from the protists Acanthamoeba castellanii and Euglena

gracilis. Gene, 44, 63-70.

Hahn, T. W., O'brien, T. P., Sah, W. J. & Kim, J. H. 1998. Acridine orange staining

for rapid diagnosis of Acanthamoeba keratitis. Japanese Journal of

Ophthalmology, 42, 108-14.

References 143

Hall, T. A. 1999. BioEdit: a user-friendly biological sequence alignment editor and

analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series,

41, 95–98.

He, Y., Niederkorn, J. Y., Mcculley, J. P., Stewart, G. L., Meyer, D. R., Silvany, R.

& Dougherty, J. 1990. In vivo and in vitro collagenolytic activity of

Acanthamoeba castellanii. Investigative Ophthalmology and Visual Science,

31, 2235-2240.

Heaselgrave, W., Patel, N., Kilvington, S., Kehoe, S. C. & Mcguigan, K. G. 2006.

Solar disinfection of poliovirus and Acanthamoeba polyphaga cysts in water

- a laboratory study using simulated sunlight. Letters in Applied


Heussen, C. & Dowdle, E. B. 1980. Electrophoretic analysis of plasminogen

activators in polyacrylamide gels containing sodium dodecyl sulphate and

copolymerized substrates. Analytical Biochemistry, 102, 196-202.

Hewett, M. K. 2003. Identification of a New Acanthamoeba 18S rRNA Gene

Sequence Type, Corresponding to the Species Acanthamoeba jacobsi

Sawyer, Nerad and Visvesvara, 1992 (Lobosea: Acanthamoebidae). Acta

Protozoologica, 42, 325-329.

Horn, M., Fritsche, T. R., Gautom, R. K., Schleifer, K. H. & Wagner, M. 1999.

Novel bacterial endosymbionts of Acanthamoeba spp. related to the

Paramecium caudatum symbiont Caedibacter caryophilus. Environmental


Horn, M. & Wagner, M. 2004. Bacterial endosymbionts of free-living amoebae.


Hu, Q. & Henney, H. R., Jr. 1997. An Acanthamoeba polyubiquitin gene and

application of its promoter to the establishment of a transient transfection

system. Biochimica et Biophysica Acta, 1351, 126-136.

Huang, S. W. & Hsu, B. M. 2010. Isolation and identification of Acanthamoeba

from Taiwan spring recreation areas using culture enrichment combined

with PCR. Acta Tropica, 115, 282-7.

Hughes, R., Heaselgrave, W. & Kilvington, S. 2003. Acanthamoeba polyphaga

strain age and method of cyst production influence the observed efficacy of

therapeutic agents and contact lens disinfectants. Antimicrobial Agents and

Chemotherapy, 47, 3080-4.

Hughes, R. & Kilvington, S. 2001. Comparison of Hydrogen Peroxide Contact Lens

Disinfection Systems and Solutions against Acanthamoeba polyphaga.

Antimicrobial Agents and Chemotherapy, 45, 2038-2043.

Hurt, M., Neelam, S., Niederkorn, J. Y. & Alizadeh, H. 2003. Pathogenic

Acanthamoeba spp. secrete a mannose-induced cytolytic protein that

correlates with the ability to cause disease. Infection and Immunity, 71,

6243-6255.

Huws, S. A., Mcbain, A. J. & Gilbert, P. 2005. Protozoan grazing and its impact

upon population dynamics in biofilm communities. Journal of Applied


Inoue, H. 1990. High efficiency transformation of Escherichia coli with plasmids.

Gene, 96, 23-28.

Inoue, T., Asari, S., Tahara, K., Kiritoshi, A., Inoue, Y. & Shimomura, Y. 1999.

Utility of Fungiflora Y stain in rapid diagnosis of Acanthamoeba keratitis.

British Journal of Ophthalmology, 83, 628-633.

References 144

Invitrogen. 2007. Plasma and Serum Preparation [Online]. Available:

http:www.invitrogen.com/site/us/en/home/References/protocols/cell-and-

tissue-analysis/elisa-protocol/ELISA-Sample-Preparation-Protocols/Plasma-

and-Serum-Preparation.html

[Accessed 14/03/2012].

Jager, B. V. & Stamm, W. P. 1972. Brain abscesses caused by free-living amoeba

probably of the genus Hartmannella in a patient with Hodgkin's disease.

Lancet, 2, 1343-1345.

Jahnes, W. G., Fullmer, H. & Li, C. P. 1957. Free living amoeba as contaminants in

monkey kidney tissue culture. Proceedings of the Society for Experimental

Biology and Medicine, 96, 484.

Jantzen, H., Schulze, I. & Stohr, M. 1990. The importance of mitotic control for

establishment of developmental competence in Acanthamoeba castellanii.

Journal of Cell Science, 97, 715-724.

Jokiranta, T. S., Jokipii, L. & Meri, S. 1995. Complement resistance of parasites.

Scandinavian Journal of Immunology, 42, 9-20.

Jones, D. B., Visvesvara, G. S. & Robinson, N. M. 1975. Acanthamoeba polyphaga

keratitis and Acanthamoeba uveitis associated with fatal

meningoencephalitis. Transactions of the Ophthalmological Societies of the

UK, 95, 221-32.

Kadlec, V. 1978. The occurrence of amphizoic Amebae in domestic animals.

Journal of protozoology, 25, 235-237.

Khan, N. A. 2006. Acanthamoeba: biology and increasing importance in human

health. FEMS Microbiology Reviews, 30, 564-595.

Khan, N. A. 2009. Acanthamoeba: Biology and Pathogenesis, Norfolk, Caister

Academic Press.

Khan, N. A., Jarroll, E. L. & Paget, T. A. 2001. Acanthamoeba can be differentiated

clinically by the polymerase chain reaction and simple plate assays. Current


Khan, N. A., Jarroll, E. L. & Paget, T. A. 2002. Molecular and physiological

differentiation between pathogenic and nonpathogenic Acanthamoeba.

Current Microbiology, 45, 197-202.

Khan, N. A., Jarroll, E. L., Panjwani, N., Cao, Z. & Paget, T. A. 2000. Proteases as

Markers for Differentiation of Pathogenic and Nonpathogenic Species of

Acanthamoeba. Journal of Clinical Microbiology, 38, 2858-2861.

Kilvington, S. 1991. Moist-heat disinfection of Acanthamoeba cysts. Reviews of

Infectious Diseases, 13 Suppl 5, S418.

Kilvington, S., Beeching, J. R. & White, D. G. 1991. Differentiation of

Acanthamoeba strains from infected corneas and the environment by using

restriction endonuclease digestion of whole-cell DNA. Journal of Clinical


Kilvington, S., Gray, T., Dart, J., Morlet, N., Beeching, J. R., Frazer, D. G. &

Matheson, M. 2004. Acanthamoeba keratitis: the role of domestic tap water

contamination in the United Kingdom. Investigative Ophthalmology and

Visual Science, 45, 165-169.

Kilvington, S., Hughes, R., Byas, J. & Dart, J. 2002. Activities of therapeutic agents

and myristamidopropyl dimethylamine against Acanthamoeba isolates.


Kilvington, S. & White, D. G. 1994. Acanthamoeba: biology, ecology and human

disease. Reviews in Medical Microbiology, 5, 12-26.

http://www.invitrogen.com/site/us/en/home/References/protocols/cell-and-tissue-analysis/elisa-protocol/ELISA-Sample-Preparation-Protocols/Plasma-and-Serum-Preparation.html



References 145

Kinde, H., Read, D. H., Daft, B. M., Manzer, M., Nordhausen, R. W., Kelly, D. J.,

Fuerst, P. A., Booton, G. & Visvesvara, G. S. 2007. Infections caused by

pathogenic free-living amebas (Balamuthia mandrillaris and Acanthamoeba

sp.) in horses. Journal of Veterinary Diagnostic Investigation, 19, 317-322.

Klopocka, W., Redowicz, M. J. & Wasik, A. 2009. Regulation of cortical

cytoskeleton dynamics during migration of free-living amoebae. Postepy

Biochemii, 55, 129-137.

Kohsler, M., Leitner, B., Blaschitz, M., Michel, R., Aspock, H. & Walochnik, J.

2006. ITS1 sequence variabilities correlate with 18S rDNA sequence types

in the genus Acanthamoeba (Protozoa: Amoebozoa). Parasitology Research,

98, 86-93.

Kohsler, M., Leitsch, D., Duchene, M., Nagl, M. & Walochnik, J. 2009.

Acanthamoeba castellanii: growth on human cell layers reactivates

attenuated properties after prolonged axenic culture. FEMS Microbiology

Letters, 299, 121-127.

Kohsler, M., Leitsch, D., Furnkranz, U., Duchene, M., Aspock, H. & Walochnik, J.

2008. Acanthamoeba strains lose their abilities to encyst synchronously

upon prolonged axenic culture. Parasitology Research, 102, 1069-72.

Kong, H. H. 2009. Molecular phylogeny of Acanthamoeba. Korean Journal of

Parasitology, 47 Suppl, S21-8.

Kuashal, V., Chhina, D. K., Kumar, R., Pannu, H. S., Dhooria, H. P. S. & Chhina,

R. S. 2008. Acanthamoeba encephalitis. Indian Journal of Medical


Kudryavtsev, A., Bernhard, D., Schlegel, M., Chao, E. E. & Cavalier-Smith, T.

2005. 18S ribosomal RNA gene sequences of Cochliopodium

(Himatismenida) and the phylogeny of Amoebozoa. Protist, 156, 215-24.

Kuiper, M. W., Valster, R. M., Wullings, B. A., Boonstra, H., Smidt, H. & Van Der

Kooij, D. 2006. Quantitative detection of the free-living amoeba

Hartmannella vermiformis in surface water by using real-time PCR. Applied

and Environmental Microbiology, 72, 5700.

Kumar, S., Tamura, K. & Nei, M. 2004. MEGA3: Integrated software for Molecular

Evolutionary Genetics Analysis and sequence alignment. Briefings in

Bioinformatics, 5.

Kurtzman, C. P. & Robnett, C. J. 2007. Multigene phylogenetic analysis of the

Trichomonascus, Wickerhamiella and Zygoascus yeast clades, and the

proposal of Sugiyamaella gen. nov. and 14 new species combinations.

FEMS Yeast Research, 7, 141-51.

Lackner, P., Beer, R., Broessner, G., Helbok, R., Pfausler, B., Brenneis, C., Auer,

H., Walochnik, J. & Schmutzhard, E. 2010. Acute granulomatous

acanthamoeba encephalitis in an immunocompetent patient. Neurocritical

Care, 12, 91-4.

Lai, S. & Henney, H. R., Jr. 1993. Nucleotide sequence of a portion of the 26S

rDNA of Acanthamoeba. Nucleic Acids Research, 21, 4401.

Larkin M.A., Blackshields G., Brown N.P., Chenna R., Mcgettigan P.A., Mcwilliam

H., Valentin F., Wallace I.M., Wilm A., Lopez R., Thompson J.D., Gibson

T.J. & D.G., H. 2007. ClustalW and ClustalX version 2. Bioinformatics 23,

2947-2948.

Ledee, D. R., Booton, G. C., Awwad, M. H., Sharma, S., Aggarwal, R. K., Niszl, I.

A., Markus, M. B., Fuerst, P. A. & Byers, T. J. 2003. Advantages of using

mitochondrial 16S rDNA sequences to classify clinical isolates of

References 146

Acanthamoeba. Investigative Ophthalmology and Visual Science, 44, 1142-

1149.

Ledee, D. R., Hay, J., Byers, T. J., Seal, D. V. & Kirkness, C. M. 1996.

Acanthamoeba griffini. Molecular characterization of a new corneal

pathogen. Investigative Ophthalmology and Visual Science, 37, 544-550.

Ledee, D. R., Seal, D. V. & Byers, T. J. 1998. Confirmatory evidence from 18S

rRNA gene analysis for in vivo development of propamidine resistance in a

temporal series of Acanthamoeba ocular isolates from a patient.


Lee, W. B. & Gotay, A. 2010. Bilateral Acanthamoeba keratitis in Synergeyes

contact lens wear: clinical and confocal microscopy findings. Eye Contact

Lens, 36, 164-9.

Leher, H., Zaragoza, F., Taherzadeh, S., Alizadeh, H. & Niederkorn, J. Y. 1999.

Monoclonal IgA Antibodies Protect Against Acanthamoeba Keratitis.

Experimental Eye Research, 69, 75-84.

Lesh, F. A. 1975. Massive development of amebas in the large intestine. Fedor

Aleksandrovich Lesh (Losch). American Journal of Tropical Medicine and

Hygiene, 24, 383-392.

Levchenko, A. & Iglesias, P. A. 2002. Models of Eukaryotic Gradient Sensing:

Application to Chemotaxis of Amoebae and Neutrophils. Biophysical

Journal, 82, 50-63.

Lim, L., Coster, D. J. & Badenoch, P. R. 2000. Antimicrobial susceptibility of 19

Australian corneal isolates of Acanthamoeba. Clinical and Experimental


Liu, H., Moon, E. K., Yu, H. S., Jeong, H. J., Hong, Y. C., Kong, H. H. & Chung,

D. I. 2005. Evaluation of taxonomic validity of four species of

Acanthamoeba: A. divionensis, A. paradivionensis, A. mauritaniensis, and A.

rhysodes, inferred from molecular analyses. Korean Journal of


Lorenzo-Morales, J., Monteverde-Miranda, C. A., Jimenez, C., Tejedor, M. L.,

Valladares, B. & Ortega-Rivas, A. 2005. Evaluation of Acanthamoeba

isolates from environmental sources in Tenerife, Canary Islands, Spain.

Annals of Agricultural and Environmental Medicine, 12, 233-6.

Ly, T. M. & Muller, H. E. 1990. Ingested Listeria monocytogenes survive and

multiply in protozoa. Journal of medical microbiology, 33, 51-54.

Ma, P., Visvesvara, G., Martinez, A. J., Theodore, F. H., Daggett, P. M. & Sawyer,

T. K. 1990. Naegleria and Acanthamoeba infections: Review. Reviews of

Infectious Diseases, 12, 490-513.

Maghsood, A. H., Sissons, J., Rezaian, M., Nolder, D., Warhurst, D. & Khan, N. A.

2005. Acanthamoeba genotype T4 from the UK and Iran and isolation of the

T2 genotype from clinical isolates. Journal of Medical Microbiology, 54,

755-759.

Marciano-Cabral, F. 2003. Acanthamoeba spp. as agents of disease in humans.

Clinical Microbiology Reviews, 16, 273-307.

Martinez, A. J. 1980. Is Acanthamoeba encephalitis an opportunistic infection?

Neurology, 30, 567-574.

Martinez, A. J. 1997. Free-living, amphizoic and opportunistic amebas. Brain

Pathology, 7, 583-598.

Martinez, A. J. & Janitschke, K. 1985. Acanthamoeba, an opportunistic

microorganism: a review. Infection, 13, 251-256.

References 147

Martinez, A. J. & Visvesvara, G. S. 1997. Free-living, amphizoic and opportunistic

amebas. Brain pathology (Zurich, Switzerland), 7, 583-598.

Matoba, A. Y., Pare, P. D., Le, T. D. & Osato, M. S. 1989. The effects of freezing

and antibiotics on the viability of Acanthamoeba cysts. Archives of


Matson, D. O., Rouah, E., Lee, R. T., Armstrong, D., Parke, J. T. & Baker, C. J.

1988. Acanthameba meningoencephalitis masquerading as

neurocysticercosis. Pediatric Infectious Disease Journal, 7, 121-4.

Mattana, A., Biancu, G., Alberti, L., Accardo, A., Delogu, G., Fiori, P. L. &

Cappuccinelli, P. 2004. In vitro evaluation of the effectivness of the

macrolide, rokitamycin and chlorpromazine against Acanthamoeba

castellanii. Antimicrobial Agents and Chemotherapy, 48, 4520-4527.

May, L. P., Sidhu, G. S. & Buchness, M. R. 1992. Diagnosis of Acanthamoeba

infection by cutaneous manifestations in a man seropositive to HIV. Journal

of the American Academy of Dermatology, 26, 352-5.

Mazur, T., Hadas, E. & Iwanicka, I. 1995. The duration of the cyst stage and the

viability and virulence of Acanthamoeba isolates. Tropical medicine and

parasitology : official organ of Deutsche Tropenmedizinische Gesellschaft

and of Deutsche Gesellschaft fur Technische Zusammenarbeit (GTZ), 46,

106-108.

Mckellar, M. S., Mehta, L. R., Greenlee, J. E., Hale, D. C., Booton, G. C., Kelly, D.

J., Fuerst, P. A., Sriram, R. & Visvesvara, G. S. 2006. Fatal granulomatous

Acanthamoeba encephalitis mimicking a stroke, diagnosed by correlation of

results of sequential magnetic resonance imaging, biopsy, in vitro culture,

immunofluorescence analysis, and molecular analysis. Journal of Clinical


Meisler, D. M., Ludwig, I. H., Rutherford, I., Bican, F. E., Langston, R. H. &

Visvesvara, G. S. 1986. Susceptibility of Acanthamoeba to cryotherapeutic

method. Archives of Ophthalmology, 104, 130-131.

Mietz, H. & Font, R. L. 1997. Acanthamoeba keratitis with granulomatous reaction

involving the stroma and anterior chamber. Archives of Ophthalmology, 115,

259-63.

Mogoa, E., Bodet, C., Legube, B. & Hechard, Y. 2010. Acanthamoeba castellanii:

Cellular changes induced by chlorination. Experimental Parasitology, In

press.

Molmeret, M., Horn, M., Wagner, M., Santic, M. & Abu Kwaik, Y. 2005. Amoebae

as training grounds for intracellular bacterial pathogens. Applied and

Environmental Microbiology, 71, 20-28.

Moore, M. B., Mcculley, J. P., Newton, C., Cobo, L. M., Foulks, G. N., O'day, D.

M., Johns, K. J., Driebe, W. T., Wilson, L. A., Epstein, R. J. & Et Al. 1987.

Acanthamoeba keratitis. A growing problem in soft and hard contact lens

wearers. Ophthalmology, 94, 1654-61.

Moura, H., Wallace, S. & Visvesvara, G. S. 1992. Acanthamoeba healyi N. Sp. and

the Isoenzyme and Immunoblot Profiles of Acanthamoeba spp., Groups 1

and 3. Journal of Eukaryotic Microbiology, 39, 573-583.

Mrva, M., Garajova, M., Lukac, M. & Ondriska, F. 2011. Weak cytotoxic activity

of miltefosine against clinical isolates of Acanthamoeba spp. Journal of


Murdoch, D., Gray, T. B., Cursons, R. & Parr, D. 1998. Acanthamoeba keratitis in

New Zealand, including two cases with in vivo resistance to

References 148

polyhexamthylene biguanide. Australian and New Zealand Journal of

Opthalmology, 26, 231-236.

Na, B. K., Kim, J. C. & Song, C. Y. 2001. Characterisation and pathogenic role of a

proteinase from Acanthamoeba castellanii. Microbial Pathogenesis, 30, 39-

48.

Nachega, J. B., Rombaux, P., Weynand, B., Thomas, G. & Zech, F. 2005.

Successful treatment of acanthamoeba rhinosinusitis in a patient with AIDS.

AIDS Patient Care, 19, 621-625.

Nassonova, E., Smirnov, A., Fahrni, J. & Pawlowski, J. 2010. Barcoding amoebae:

comparison of SSU, ITS and COI genes as tools for molecular identification

of naked lobose amoebae. Protist, 161, 102-115.

Neff, R. J. 1958. Mechanisms of Purifying Amoebae by Migration on Agar

Surfaces. Journal of Eukaryotic Microbiology, 5, 226-231.

Neff, R. J., Ray, S. A., Benton, W. F. & Wilborn, M. 1964. Induction of

synchronous encystment (differentiation) in Acanthamoeba sp. Methods in

Cell Physiology, 1, 55-83.

Nellen, W. & Gallwitz, D. 1982. Actin genes and actin messenger RNA in

Acanthamoeba castellanii. Nucleotide sequence of the split actin gene I.

Journal of Molecular Biology, 159, 1-18.

Nelson, J. E. 1992. Purification of cloned and genomic DNA by guanidine

thiocyanate/isobutyl alcohol fractionation. Analytical Biochemistry, 207,

197-201.

Niederkorn, J. Y., Alizadeh, H., Leher, H. & Mcculley, J. P. 1999. The pathogenesis

of Acanthamoeba keratitis. Microbes and Infection, 1, 437-443.

Niszl, I. A., Veale, R. B. & Markus, M. B. 1998. Cytopathogenicity of clinical and

environmental Acanthamoeba isolates for two mammalian cell lines.

Journal of Parasitology, 84, 961-967.

Niyyati, M., Lorenzo-Morales, J., Rezaie, S., Rahimi, F., Mohebali, M., Maghsood,

A. H., Motevalli-Haghi, A., Martin-Navarro, C. M., Farnia, S., Valladares,

B. & Rezaeian, M. 2009. Genotyping of Acanthamoeba isolates from

clinical and environmental specimens in Iran. Experimental Parasitology,

121, 242-5.

Obaray, N. & Coakley, W. T. 2001. Phagocytosis by Acanthamoeba castellanii:

ionic strength dependence of the probability of cell attachment; ingestion

and contact seam morphology. Colloids and surfaces.B, Biointerfaces, 22,

127-140.

Ofori-Kwakye, S. K., Sidebottom, D. G., Herbert, J., Fischer, E. G. & Visvesvara,

G. S. 1986. Granulomatous brain tumor caused by Acanthamoeba. Case

report. Journal of Neurosurgery, 64, 505-9.

Page, F. C. 1967. Re-definition of the genus Acanthamoeba with descriptions of

three species. Journal of Protozoology, 14, 709-724.

Page, F. C. 1988. A New key to Freshwater and Soil Gymnamoebae, The Ferry

House, Ambleside, Cumbria, UK, Freshwater Biological Association.

Pearce, J. R., Powell, H. S., Chandler, F. W. & Visvesvara, G. S. 1985. Amebic

meningoencephalitis caused by Acanthamoeba castellani in a dog. Journal

of the American Veterinary Medical Association, 187, 951-952.

Perez-Santonja, J. J., Kilvington, S., Hughes, R., Tufail, A., Matheson, M. & Dart,

J. 2003. Persistently culture positive acanthamoeba keratitis: in vivo

resistance and in vitro sensitivity. Opthalmology, 110, 1593-1600.

References 149

Perrine, D., Chenu, J. P., Georges, P., Lancelot, J. C., Saturnino, C. & Robba, M.

1995. Amoebicidal efficiencies of various diamidines against two strains of

Acanthamoeba polyphaga. Antimicrobial Agents and Chemotherapy, 39,

339-342.

Pettit, D. A., Williamson, J., Cabral, G. A. & Marciano-Cabral, F. 1996. In vitro

destruction of nerve cell cultures by Acanthamoeba spp.: a transmission and

scanning electron microscopy study. Journal of Parasitolology, 82, 769-77.

Pollard, T. D. 1981. Cytoplasmic contractile proteins. Journal of Cell Biology, 91,

156s-165s.

Pollard, T. D. & Rimm, D. L. 1991. Analysis of cDNA clones for Acanthamoeba

profilin-I and profilin-II shows end to end homology with vertebrate

profilins and a small family of profilin genes. Cell Motility and the

Cytoskeleton, 20, 169-77.

Prusch, R. D. & Roscoe, J. C. 1993. A possible signal-coupling role for cyclic AMP

during endocytosis in Amoeba proteus. Tissue and Cell, 25, 141-149.

Pumidonming, W., Kohsler, M. & Walochnik, J. 2010. Acanthamoeba strains show

reduced temperature tolerance after long-term axenic culture. Parasitology

Research, 106, 553-9.

Pumidonming, W., Walochnik, J., Dauber, E. & Petry, F. 2011. Binding to

complement factors and activation of the alternative pathway by

Acanthamoeba. Immunobiology, 216, 225-33.

Puschkarew, B. 1913. Über die Verbreitung der Süsswasser protozoen durch die

Luft. . Arch. Protistent., 23.

Pussard, M. & Pons, R. 1977. Morphologie de la paroi kystiqueet taxonomie du

genre Acanthamoeba (Protozoa, Amoebida). Protistologica, 8, 557-598.

Qvarnstrom, Y., Da Silva, A. J., Schuster, F. L., Gelman, B. B. & Visvesvara, G. S.

2009. Molecular confirmation of Sappinia pedata as a causative agent of

amoebic encephalitis. Journal of Infectious Diseases, 199, 1139-1142.

Qvarnstrom, Y., Visvesvara, G. S., Sriram, R. & Da Silva, A. J. 2006. Multiplex

Real-Time PCR Assay for Simultaneous Detection of Acanthamoeba spp.,

Balamuthia mandrillaris, and Naegleria fowleri. Journal of Clinical


Radford, C. F., Bacon, A. S., Dart, J. K. & Minassian, D. C. 1995. Risk factors for

acanthamoeba keratitis in contact lens users: a case-control study. British

Medical Journal, 310, 1567-70.

Radford, C. F., Lehmann, O. J. & Dart, J. K. 1998. Acanthamoeba keratitis:

multicentre survey in England 1992-6. National Acanthamoeba Keratitis

Study Group. British Journal of Ophthalmology, 82, 1387-92.

Radford, C. F., Minassian, D. C. & Dart, J. K. 2002. Acanthamoeba keratitis in

England and Wales: incidence, outcome, and risk factors. British Journal of


Raederstorff, D. & Rohmer, M. 1985. Sterol biosynthesis de novo via cycloartenol

by the soli amoebae Acanthamoeba polyphaga. Biochemical Journal, 231,

609-615.

Ren, M. & Wu, X. 2010. Evaluation of three different methods to establish animal

models of Acanthamoeba keratitis. Yonsei Medical Journal, 51, 121-127.

Rimm, D. L., Pollard, T. D. & Hieter, P. 1988. Resolution of Acanthamoeba

castellanii chromosomes by pulsed field gel electrophoresis and construction

of the initial linkage map. Chromosoma, 97, 219-23.

References 150

Riviere, D., Szczebara, F. M., Berjeaud, J. M., Frere, J. & Hechard, Y. 2006.

Development of a real-time PCR assay for quantification of Acanthamoeba

trophozoites and cysts. Journal of Microbiological Methods, 64, 78-83.

Rodriguez-Zaragoza, S. 1994. Ecology of free-living amoebae. Critical Reviews in


Rowbotham, T. J. 1998. Isolation of Legionella pneumophila serogroup 1 from

human feces with use of amebic cocultures. Clinical Infectious Diseases, 26,

502-503.

Sambrook, J., Fritish, E. F. & Maniatis, T. 1989. Molecular Cloning: A Laboratory

Manual, New York, USA, Cold Spring Harbor Laboratory Press.

Sanati, H., Belanger, P., Fratti, R. & Ghannoum, M. 1997. A new triazole,

voriconazole (UK-109,496), blocks sterol biosynthesis in Candida albicans

and Candida krusei. Antimicrobial Agents and Chemotherapy, 41, 2492-

2496.

Sawyer, T. K. 1971. Acanthamoeba griffini, a New Species of Marine Amoeba.


Sawyer, T. K. 1989. Free-living pathogenic and nonpathogenic amoebae in

Maryland soils. Applied and Environmental Microbiology, 55, 1074-1077.

Schaumberg, D. A., Snow, K. K. & Dana, M. R. 1998. The epidemic of

Acanthamoeba keratitis: where do we stand? Cornea, 17, 3-10.

Schroeder-Diedrich, J. M., Fuerst, P. A. & Byers, T. J. 1998. Group-I introns with

unusual sequences occur at three sites in nuclear 18S rRNA genes of

Acanthamoeba lenticulata. Current Genetics, 34, 71-78.

Schroeder, J. M., Booton, G. C., Hay, J., Niszl, I. A., Seal, D. V., Markus, M. B.,

Fuerst, P. A. & Byers, T. J. 2001. Use of subgenic 18S ribosomal DNA PCR

and sequencing for genus and genotype identification of acanthamoebae

from humans with keratitis and from sewage sludge. Journal of Clinical


Schuster, F. L. 2002. Cultivation of pathogenic and opportunistic free-living

amebas. Clinical Microbiology Reveiws, 15, 342-54.

Schuster, F. L., Guglielmo, B. J. & Visvesvara, G. S. 2006a. In-vitro activity of

miltefosine and voriconazole on clinical isolates of free-living amebas:

Balamuthia mandrillaris, Acanthamoeba spp., and Naegleria fowleri.


Schuster, F. L., Honarmand, S., Visvesvara, G. S. & Glaser, C. A. 2006b. Detection

of antibodies against free-living amoebae Balamuthia mandrillaris and

Acanthamoeba species in a population of patients with encephalitis. Clinical

Infectious Diseases, 42, 1260-5.

Schuster, F. L. & Visvesvara, G. S. 1998. Efficacy of novel antimicrobials against

clinical isolates of opportunistic amebas. Journal of Eukaryotic


Schuster, F. L. & Visvesvara, G. S. 2004. Free-living amoebae as opportunistic and

non-opportunistic pathogens of humans and animals. International Journal

of Parasitology, 34, 1001-1027.

Seal, D. V. 2003. Acanthamoeba keratitis update-incidence, molecular

epidemiology and new drugs for treatment. Eye, 17, 893-905.

Seijo Martinez, M., Gonzalez-Mediero, G., Santiago, P., Rodriguez De Lope, A.,

Diz, J., Conde, C. & Visvesvara, G. S. 2000. Granulomatous amebic

encephalitis in a patient with AIDS: isolation of acanthamoeba sp. Group II

References 151

from brain tissue and successful treatment with sulfadiazine and fluconazole.

Journal of Clinical Microbiology, 38, 3892-5.

Sharma, P. P., Gupta, P., Murali, M. V. & Ramachandran, V. G. 1993. Primary

amebic meningoencephalitis caused by Acanthamoeba: successfully treated

with cotrimoxazole. Indian Pediatrics, 30, 1219-22.

Sharma, S., Ramachandran, L. & Rao, G. N. 1995. Adherence of cysts and

trophozoites of Acanthamoeba to unworn rigid gas permeable and soft

contact lenses. Contact Lens Association of Ophalmologists Journal, 21,

247-51.

Sinard, J. H., Stafford, W. F. & Pollard, T. D. 1989. The mechanism of assembly of

Acanthamoeba myosin-II minifilaments: minifilaments assemble by three

successive dimerization steps. Journal of Cell Biology, 109, 1537-1547.

Singhal, T., Bajpai, A., Kalra, V., Kabra, S. K., Samantaray, J. C., Satpathy, G. &

Gupta, A. K. 2001. Successful treatment of Acanthamoeba meningitis with

combination oral antimicrobials. The Pediatric Infectious Disease Journal,

20, 623-627.

Sissons, J., Alsam, S., Goldsworthy, G., Lightfoot, M., Jarroll, E. L. & Khan, N. A.

2006. Identification and properties of proteases from an Acanthamoeba

isolate capable of producing granulomatous encephalitis. BMC

Microbiology, 6, 42.

Slater, C. A., Sickel, J. Z., Visvesvara, G. S., Pabico, R. C. & Gaspari, A. A. 1994.

Successful treatment of disseminated Acanthamoeba infection in an

immunocompromised patient. New England Journal of Medicine, 331,

85=87.

Smirnov, A. V., Nassonova, E. S., Chao, E. & Cavalier-Smith, T. 2007. Phylogeny,

evolution, and taxonomy of vannellid amoebae. Protist, 158, 295-324.

Srinivasan, M., Channa, P., Raju, C. V. & George, C. 1993. Acanthamoeba keratitis

in hard contact lens wearer. Indian Journal of Ophthalmology, 41, 187-8.

Sriram, R., Shoff, M., Booton, G., Fuerst, P. & Visvesvara, G. S. 2008. Survival of

Acanthamoeba cysts after desiccation for more than 20 years. Journal of

clinical microbiology, 46, 4045-4048.

Steenbergen, J. N., Shuman, H. A. & Casadevall, A. 2001. Cryptococcus

neoformans interactions with amoebae suggest an explanation for its

virulence and intracellular pathogenic strategy in macrophages. Proceedings

of the National Academy of Sciences of the United States of America, 98,

15245-15250.

Stehr-Green, J. K., Bailey, T. M. & Visvesvara, G. S. 1989. The epidemiology of

Acanthamoeba keratitis in the United States. American Journal of


Stothard, D. R., Hay, J., Schroeder-Diedrich, J. M., Seal, D. V. & Byers, T. J. 1999.

Fluorescent oligonucleotide probes for clinical and environmental detection

of Acanthamoeba and the T4 18S rRNA gene sequence type. Journal of

Clinical Microbiology, 37, 2687-2693.

Stothard, D. R., Schroeder-Diedrich, J. M., Awwad, M., Gast, R. J., Ledee, D. R.,

Rodriguez-Zaragoza, S. D. C. L., Fuerst, P. A. & Byers, T. 1998. The

evolutionary history of the genus Acanthamoeba and the identification of

eight new 18S rRNA gene sequence types. Journal of Eukaryotic


References 152

Stratford, M. P. & Griffiths, A. J. 1978. Variations in the Properties and

Morphology of Cysts of Acanthamoeba castellanii. Journal of General


Tamura, K., Nei, M. & Kumar, S. 2004. Prospects for inferring very large

phylogenies by using the neighbor-joining method. Proceedings of the

National Academy of Sciences (USA), 101, 11030-11035.

Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M. & Kumar, S. 2011.

MEGA5: Molecular Evolutionary Genetics Analysis using Maximum

Likelihood, Evolutionary Distance, and Maximum Parsimony Methods.

Molecular Biology and Evolution, In Press.

Taylor, D. L. 1977. The contractile basis of amoeboid movement : IV. The

viscoelasticity and contractility of amoeba cytoplasm in vivo. Experimental

Cell Research, 105, 413-426.

Thamprasert, K., Khunamornpong, S. & Morakote, N. 1993. Acanthamoeba

infection of peptic ulcer. Annals of Tropical Medicine and Parasitology, 87,

403-5.

Thomas, V., Herrera-Rimann, K., Blanc, D. S. & Greub, G. 2006. Biodiversity of

amoebae and amoeba-resisting bacteria in a hospital water network. Applied

and Environmental Microbiology, 72, 2428-38.

Toney, M. D. & Marciano-Cabral, F. 1998. Resistance of Acanthamoeba species to

complement lysis. Journal of Parasitology, 84, 338-344.

Torno, M. S., Jr., Babapour, R., Gurevitch, A. & Witt, M. D. 2000. Cutaneous

acanthamoebiasis in AIDS. Journal of the American Academy of

Dermatology, 42, 351-4.

Vaddavalli, P. K., Garg, P., Sharma, S., Sangwan, V. S., Rao, G. N. & Thomas, R.

2011. Role of confocal microscopy in the diagnosis of fungal and

acanthamoeba keratitis. Ophthalmology, 118, 29-35.

Vemuganti, G. K., Sharma, S., Athmanathan, S. & Garg, P. 2000. Keratocyte loss in

Acanthamoeba keratitis: phagocytosis, necrosis or apoptosis? Indian Journal

of Ophthalmology, 48, 291-4.

Verani, J. R., Lorick, S. A., Yoder, J. S., Beach, M. J., Braden, C. R., Roberts, J. M.,

Conover, C. S., Chen, S., Mcconnell, K. A., Chang, D. C., Park, B. J., Jones,

D. B., Visvesvara, G. S. & Roy, S. L. 2009. National outbreak of

Acanthamoeba keratitis associated with use of a contact lens solution,

United States. Emerging Infectious Diseases, 15, 1236-42.

Villalobos, N., González, L. M., Morales, J., De Aluja, A. S., Jiménez, M. I.,

Blanco, M. A., Harrison, L. J. S., Parkhouse, R. M. E. & Gárate, T. 2007.

Molecular identification of Echinococcus granulosus genotypes (G1 and

G7) isolated from pigs in Mexico. Veterinary Parasitology, 147, 185-189.

Visvesvara, G. S. 2010. Free-Living Amebae as Opportunistic Agents of Human

Disease. Journal of Neuroparasitology, 1, 1-13.

Visvesvara, G. S., Booton, G. C., Kelley, D. J., Fuerst, P., Sriram, R., Finkelstein,

A. & Garner, M. M. 2007a. In vitro culture, serologic and molecular analysis

of Acanthamoeba isolated from the liver of a keel-billed toucan

(Ramphastos sulfuratus). Veterinary Parasitology, 143, 74-78.

Visvesvara, G. S., Moura, H. & Schuster, F. L. 2007b. Pathogenic and opportunistic

free-living amoebae: Acanthamoeba spp., Balamuthia mandrillaris,

Naegleria fowleri, and Sappinia diploidea. FEMS Immunology and Medical


References 153

Vogelstein, B. & Gillespie, D. 1979. Preparative and Analytical Purification of

DNA from Agarose. Proceedings of the National Academy of Sciences, 76,

615-619.

Volkonsky, M. 1931. Hartmannella castellanii Douglas et classification des

Hartmannelles. Archives of Zoological Experimental Genetics, 72.

Wainright, P. O., Hinkle, G., Sogin, M. L. & Stickel, S. K. 1993. Monophyletic

origins of the metazoa: an evolutionary link with fungi. Science, 260, 340-2.

Walia, R., Montoya, J. G., Visvesvera, G. S., Booton, G. C. & Doyle, R. L. 2007. A

case of successful treatment of cutaneous Acanthamoeba infection in a lung

transplant recipient. Transplant Infectious Disease, 9, 51-4.

Walochnik, J., Aichelburg, A., Assadian, O., Steuer, A., Visvesvara, G., Vetter, N.

& Aspock, H. 2008. Granulomatous amoebic encephalitis caused by

Acanthamoeba amoebae of genotype T2 in a human immunodeficiency

virus-negative patient. Journal of Clinical Microbiology, 46, 338-340.

Walochnik, J., Duchene, M., Seifert, K., Obwaller, A., Hottkowitz, T., Wiedermann,

G., Eibl, H. & Aspock, H. 2002. Cytotoxic activities of

alkylphosphocholines against clinical isolates of Acanthamoeba spp.

Antimicrobial Agents and Chemotherapy, 2002, 3.

Walochnik, J., Haller-Schober, E. M., Kolli, H., Picher, O., Obwaller, A. & Aspock,

H. 2000. Discrimination between clinically relevant and nonrelevant

Acanthamoeba strains isolated from contact lens-wearing keratitis patients in

Austria. Journal of Clinical Microbiology, 38, 3932-3936.

Walochnik, J., Obwaller, A., Gruber, F., Mildner, M., Tschachler, E., Suchomel,

M., Duchene, M. & Auer, H. 2009. Anti-Acathamoeba efficacy and toxicity

of miltefosine in an organotypic skin equivalent. Journal of Antimicrobial

Chemotherapy, 64, 539-545.

Ward, A. P. & Al-Abidin, N. Z. 1988. Multiple molecular forms of Acanthamoeba

lactic dehydrogenase. Comparative Biochemistry and Physiology Part B:,

90, 833-6.

Watkins, R. F. & Gray, M. W. 2006. The frequency of eubacterium-to-eukaryote

lateral gene transfers shows significant cross-taxa variation within

amoebozoa. Journal of Molecular Evolution, 63, 801-14.

Weekers, P. H., Gast, R. J., Fuerst, P. A. & Byers, T. J. 1994. Sequence variations

in small-subunit ribosomal RNAs of Hartmannella vermiformis and their

phylogenetic implications. Molecular Biology and Evolution, 11, 684-690.

Wilhelmus, K. R., Jones, D. B., Matoba, A. Y., Hamill, M. B., Pflugfelder, S. C. &

Weikert, M. P. 2008. Bilateral acanthamoeba keratitis. American Journal of


Woese, C. R., Kandler, O. & Wheelis, M. L. 1990. Towards a natural system of

organisms: proposal for the domains Archaea, Bacteria, and Eucarya.

Proceedings of the National Academy of Sciences of the United States of

America, 87, 4576-9.

Xuan, Y. H., Hong, Y. C., Lee, Y. S., Kang, S. W., Yu, H. S., Ahn, T. I., Chung, D.

I. & Kong, H. H. 2009. Acanthamoeba healyi: expressed gene profiles with

enhanced virulence after mouse-brain passage. Experimental Parasitology,

123, 226-230.

Yin, Y. & Henney Jnr, H. R. 1997. Stable transfection of Acanthamoeba. Canadian

Journal of Microbiology, 43, 239-244.

Yoo, S. H., Romano, A. C., Alfonso, E. C., Miller, D. & Forster, R. K. 2004.

Lightening flash radial keratoneuritis in acanthamoeba keratitis visualized

References 154

with confocal microscopy. Investigative Ophthalmology and Visual Science,

45, 112-.

Yu, H., Hwang, M., Kim, T., Yun, H., Kim, T., Kong, H. & Chung, D. 1999.

Phylogenetic relationships among Acanthamoeba spp. based on PCR-RFLP

analyses of mitochondrial small subunit rRNA gene. Korean Journal of


Yu, H. S., Kong, H. H., Kim, S. Y., Hahn, Y. H., Hahn, T. W. & Chung, D. I. 2004.

Laboratory investigation of Acanthamoeba lugdunensis from patients with

keratitis. Investigative Ophthalmology and Visual Science, 45, 1418-1426.

Yue, G. H., Wang, G. L., Zhu, B. Q., Wang, C. M., Zhu, Z. Y. & Lo, L. C. 2008.

Discovery of four natural clones in a crayfish species Procambarus clarkii.

International Journal of Biological Sciences, 4, 279-82.

Zwick, M. G., Wiggs, M. & Paule, M. R. 1991. Sequence and organization of 5S

RNA genes from the eukaryotic protist Acanthamoeba castellanii. Gene,

101, 153-7.

molecular studies of the pathogenic free-living amoeba, acanthamoeba

Documents